Beruflich Dokumente
Kultur Dokumente
Full citation: McNeill, John R., and Verena Winiwarter, eds. Soils and
Societies: Perspectives from Environmental History. Cambridge:
The White Horse Press, 2010.
http://www.environmentandsociety.org/node/3489.
Rights: All rights reserved. The White Horse Press 2001. Except
for the quotation of short passages for the purpose of
criticism or review, no part of this book may be reprinted
or reproduced or utilised in any form or by any electronic,
mechanical or other means, including photocopying or
recording, or in any information storage or retrieval system,
without permission from the publishers. For further
information please see http://www.whpress.co.uk.
Soils and Societies
SOILS AND SOCIETIES
Perspectives from Environmental History
edited by
J.R. McNeill
and
Verena Winiwarter
All rights reserved. Except for the quotation of short passages for the purpose of
criticism or review, no part of this book may be reprinted or reproduced or utilised
in any form or by any electronic, mechanical or other means, including photo-
copying or recording, or in any information storage or retrieval system, without
permission from the publishers.
The inset illustration on the cover is from Wolf Helmhard von Hohberg, Georgica curiosa.
Das ist: Umstndlicher Bericht und klarer Unterricht von dem adelichen Land- und Feldleben,
Nrnberg 1682. Permission to use the copy of the Benedictine Library at Melk, Austria for
reproduction is gratefully acknowledged. Photograph Verena Winiwarter, 2005. Back-
ground photograph reproduced courtesy of Professor H.-R. Bork.
Contents
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .341
Biographical Notes
Editors
J.R. McNeill was born in Chicago and gained diplomas from Swarthmore College
and Duke University before joining the History Department and the Walsh School
of Foreign Service at Georgetown University in 1985. Since 2003 he has held the
Cinco Hermanos Chair in Environmental and International Affairs at Georgetown.
Recent books include Something New Under the Sun: An Environmental History of
the 20th-century World (New York: W.W. Norton/London: Penguin UK, 2000);
The Mountains of the Mediterranean World: An Environmental History (Cambridge:
Cambridge University Press, 1992, reissued in 2003); The Human Web: A Birds-eye
View of Human History (New York: W.W. Norton, 2003; co-authored with William
H. McNeill); and an edited volume, Environmental History in the Pacific World
(Aldershot: Ashgate, 2001). He lives an agreeably harried life with his triathlete
wife and four exuberant children, and if he had time would enjoy sports, movies
and camping.
Verena Winiwarter, who was born and still lives in Vienna, was first trained as
a chemical engineer. After years of working in atmospheric research, she earned
her M.A. in history and communication sciences in 1991, and her Ph.D. in en-
vironmental history at the University of Vienna in 1998. She has been a lecturer
(Universittsdozent) there in Human Ecology since 2003. She currently holds a
postdoctoral fellowship in environmental history (APART fellowship) awarded
by the Austrian Academy of Sciences at the Institute for Soil Research, University
of Natural Resources and Applied Life Sciences, Vienna and at the Faculty for
Interdisciplinary Research of Klagenfurt University. Her main research interests
within environmental history comprise the micro-history of landscapes, waste,
landscape images and soils. She is a founding member of the European Society for
Environmental History. From 2001 until 2005 she served as its president. Besides
work she enjoys walking and hiking and picnics with her husband and two kids.
She has published more than 80 articles and book chapters, and co-authored two
CD-ROMs. Recent book publications include Historische Humanoekologie, co-
edited with Harald Wilfing (Vienna: WUV Facultas, 2003) and the forthcoming
Das Ende der Flaeche, co-authored with Rolf Peter Sieferle et al. (Kln: Bhlau,
2006). Website: www.iff.ac.at/umweltgeschichte/winiwarter.php
viii
Biographical Notes
Authors
Andreas Mieth studied Biology, Ecology, Marine Biology and Fisheries Biol-
ogy in Berlin and Kiel (Germany). He has worked as science-coordinator in the
Centre of Biology at Kiel University from 1984 to 2001. Since 2001 he has been
the science-coordinator at the Ecology Centre and vice-head of the Department
of Ecotechnology and Ecosystem Development. His main research interests are
ecosystem and landscape development, in particular the reaction of ecosystems
under human impact; technologyecology interactions; and ecotechnology, e.g.
technical use of renewable materials, development of glues based on bio-materials,
and ecological sustainability of techniques. His Ph.D. thesis dealt with Effects of
Land Use on the Development of Landscape and Culture on Easter Island (Chile).
Publications include A. Mieth and H.-R. Bork, 2005, History, Origin and Extent
of Soil Erosion on Easter Island (Rapa Nui) Catena 63, 24460; A. Mieth and
H.-R. Bork, 2003, Diminution and Degradation of Environmental Resources
by Prehistoric Land Use on Poike Peninsula, Easter Island (Rapa Nui), Rapa Nui
Journal 17(1), 3441
Frank Uekoetter studied history, political science and the social sciences at the
universities of Freiburg and Bielefeld in Germany and the Johns Hopkins University
in Baltimore, USA. He received his Ph.D. from Bielefeld University in 2001 with
x
Biographical Notes
a dissertation on the history of air pollution control in Germany and the United
States, which was published as Von der Rauchplage zur kologischen Revolution. Eine
Geschichte der Luftverschmutzung in Deutschland und den USA 18801970 (Essen:
Klartext Verlag, 2003). In 2002, he organised the conference Nature Protection
in Nazi Germany under the auspices of the German minister for the environ-
ment Jrgen Trittin. He is currently a researcher at the Department of History
of Bielefeld University, with a project on the history of agricultural knowledge in
the twentieth century. His publications include The Frontiers of Environmental
History, a special issue of Historical Social Research (vol. 29, no. 3, 2004), and The
Green and the Brown: A History of Conservation in Nazi Germany (Cambridge and
New York: Cambridge University Press, 2006).
Robert S. Shiel lectures on soil management and climatic change at Newcastle Uni-
versity, UK. He is closely involved in research on long-term changes in vegetation
and soils, in particular on increasing diversity and improving sustainability in hay
meadows (Bardgett, Smith, Shiel, et al. Nature, in press), and manages the 108-year
old Palace Leas hay meadow, the worlds oldest grazed grassland experiment. Recent
completed archaeologically-related research has concentrated on the Mesolithic
communities of north east England (Howick) and north west Scotland (Scotlands
First Settlers), while publication of results from longer-term projects in Hungary
(Upper Tisza Project) and Greece (Boeotia Project) proceeds but slowly, though the
fieldwork is finished. At a micro level he has also investigated changes in materials
after burial and in the residual evidence of use, particularly on stone tools.
Kate Showers completed her first academic degree in anthropology, but cross-cultural
experience suggested that a knowledge of production agriculture would be more
useful. Degrees from both Michigan State University and Cornell University resulted
in a 1982 Ph.D. from Cornells Agronomy Department. Her doctoral dissertation,
involving two years of fieldwork in Lesotho, southern Africa, included studying
the language and culture as well as origins of soil erosion processes, collection of
oral environmental histories, and archival environmental history research. After
a post-doctoral fellowship in the Department of Theoretical Production Ecol-
ogy, Agricultural University, Wageningen, the Netherlands and appointments as
a research scientist at the Rodale Research Center, Pennsylvania, USA, and as an
Academic Visitor at the Social Studies Faculty Centre, Oxford University, she was
named Head of Research, Institute of Southern African Studies, National Univer-
sity of Lesotho. This enabled her to continue investigating oral historys potential
for obtaining information about historical environmental change. Currently she
is Senior Research Associate and Visiting Research Fellow at the Centre for World
Environmental History, University of Sussex (Website link: http://www.sussex.ac.uk/
development/1-4-5-2-1.html). A synthesis of field work, oral history and archival
xi
Biographical Notes
The survival of human communities has always depended, and still depends, on
certain environmental goods. Oxygen and fresh water top the list, but soil is not
far behind, because for most of history most people got most of their food from
plants grown in soils. Our distant forbears who lived by foraging and hunting de-
pended indirectly on soils, as they provided the nutrients necessary for the growth
of the fruits, nuts, tubers and, indirectly, the animals, that formed our ancestors
food supply.
When people began the transitions to agriculture (which happened multiple
times beginning about 10,000 years ago), their dependence upon soils became
more direct and more obvious to them. Agricultural peoples were less mobile than
foragers, and more likely to stay put long enough to have significant impacts upon
soils, perhaps eroding them, perhaps degrading them, perhaps enriching them.
Their religious practice, in many settings, included worship intended to maintain
the fertility of soils, something unknown among non-agricultural peoples. The
ancient Egyptians, for example, included in their pantheon a male god (Geb or
Seb) whose portfolio included the earth in general and Nile silt in particular. More
frequently, deities associated with the soil, and its fertility, were female (Winiwarter
and Blum, 2006).
With the emergence of urban life some 5,000 years ago, the relationship
between humankind and soils shifted once more. Village life had only modest
effects upon the distribution of soil nutrients, because although people harvested
and ate crops, eventually their wastes or their corpses returned the nutrients to the
local environment. But with cities things were different. Farmers exported food
to cities, and wastes and corpses never returned, or rarely returned except in small
quantities, to the fields whence the food had come. Instead, soil nutrients were
systematically drawn from the fields and, after lodging in human bodies for shorter
2
J.R. McNeill and Verena Winiwarter
or longer stays, passed into the urban environment, often to be carried away by
runoff, local streams and rivers. Henceforth, farmers who served urban markets, or
indeed distant markets of any sort, faced a structural problem of long-term nutrient
loss. Where soils were deep and rich, the problem might be ignored for generations.
Elsewhere, the problem required human ingenuity in the form of soil knowledge
and soil management. For millennia, indeed until the nineteenth century, the chief
solution to this problem took the form of animal manure. As one Polish nobleman
of the sixteenth century put it, manure is worth more than a man with a doctorate
(Gostomski 1588). By grazing animals on pasture, scrubland, or in forests, peoples
of the Old World could in effect collect nutrients from broad areas and concentrate
them on their fields, simply by penning their animals and collecting their manure.
Livestock, in addition to their other functions in agro-ecosystems, worked as nutri-
ent gatherers for farmers fields. In the New World, without large livestock before
1492, farmers lacked manure and had to invest greater labour and ingenuity if
they intended to maintain the fertility of their fields. In some areas, human wastes
(night soil) supplemented or even replaced animal manure. The use of night soil is
very ancient and (until recently) widespread: it is mentioned in Homers Odyssey
and was routine in parts of East Asia and Mesoamerica as well.
The survival, prosperity and power of almost any given farming community
rested on its success in resolving the problem of long-term nutrient loss. Egypt was
the chief exception. There the waters of the annual Nile flood brought a regular
nutrient subsidy in the form of a film of silt left behind by receding floodwaters
on the banks of the Nile. It came from the volcanic (and nutrient-rich) mountain
landscapes of Ethiopia. In effect, the Nile did the job that domestic animals (or
carts of night soil) did elsewhere. Less thoroughly and less reliably, floodwaters in
Mesopotamia and from the Indus, the Ganges, the Irrawaddy, the Mekong, the
Amazon and elsewhere served in the same capacity. Everywhere, long-term eco-
nomic trajectories, the ebb and flow of political power, the waxing and waning of
populations, rested on the successful management of soil nutrients. For want of
nitrogen, many a kingdom was lost.
This is a fundamental truth which historians, even environmental historians,
have scarcely recognised. Some inspired amateur polymaths and historically minded
soil scientists have probed the subject of humansoil relationship through time
(Hyams 1952; Lowdermilk 1953; Hillel 1991). Geographically inclined historians
have often pointed out the significance of soil factors upon human affairs, but without
acknowledging that soils have their own histories. Historians typically regard soils
as fixed features of the environment and therefore background considerations to
agricultural and economic history. The great contribution of environmental histo-
rians over the past generation is to remind others that environments are not mere
backdrops to the dramas of history, but participants in their own right, interacting
3
Soils, Soil Knowledge, and Environmental History
with all the others. This book aims to extend that logic to soils in particular, to
present them as entities with histories.
Soils have their own histories, both natural and human. Geological substrate,
climate and vegetation shape their natural histories, which in turn shape their
human histories. The human histories of soil are both material and intellectual.
What happens with soils and the societies that use them is a matter of both bio-
geochemistry and culture. What people believe about soils influences (although it
does not necessarily determine) what they do with them, whether they conserve
and nurture them, whether they abuse and abandon them. What people understand
and misunderstand about soils is thus a necessary part of any history of the
nexus between soil and society. Hence this book accords considerable attention to
soil knowledge and its maintenance, transmission and impacts, from the ancient
Roman agronomic writers to Sotho farmers and German agricultural experts.
This book also devotes considerable attention to soil erosion. Since the
1930s soil erosion has (to fluctuating degrees) occupied many a fine scholarly
mind (e.g. Neboit 1983; Blaikie and Brookfield 1985). Erosions corollary, soil
deposition, has occupied rather fewer minds. But both processes have helped shape
human affairs wherever food came from the soil. By and large scholarly discus-
sion of soil movement in history is conducted between two poles, which we may
call Malthusian and Boserupian. The Malthusian position relates soil erosion to
population pressure, growing food demand, deforestation and the cultivation of
marginal, highly erodible lands. The Boserupian pole is occupied by those who see
soil stability deriving from active efforts at conservation and enhancement, which
in the past were normally labour-intensive (e.g., building and maintaining terraces
or manuring) and thus more easily undertaken in thickly populated landscapes.
Either position might be correct in a given situation. But of course population
is not the only variable that affects patterns of erosion and deposition, often not
even the crucial one. Technologies, agricultural customs, climate change, livestock
behaviour and many other factors shape soil erosion histories. On different time
scales, different variables may dominate.
For these and other reasons, this book includes studies on different time
scales. It is organised roughly according to the length of time under consideration.
Papers deal with millennial, secular or decadal time scales. Soil creep, which typi-
cally carries small amounts of soil downslope annually, is normally a negligible
concern at decadal time scales, but sometimes a decisive one on millennial scales.
Conversely, government policies may be quite influential on decadal scales, but of
negligible impact on millennial scales (because they, like the governments behind
them, are almost never maintained long enough).
Fernand Braudel, a geographically inclined historian (although he was not
interested in soil change or soil processes), divided historical time into three parts
4
J.R. McNeill and Verena Winiwarter
in his great work on the Mediterranean world (Braudel 1972). He wrote of lhistoire
vnementielle, of la conjuncture, and of la longue dure. His categories are of shorter
duration than ours, and less arbitrary: la conjuncture for example was connected to
the 50- or 55-year Kondratieff cycles sometimes used by economic historians. But
in (faintly) echoing Braudel with our arrangement of time, beginning with longer
scales and moving to shorter ones, we are merely trying to make clear, as he was, that
different processes appear salient depending on the depth of historical vision.
Spatial scale and geographic location are also variables that affect both soil
processes and what looks important about them. Chapters in this book range in
spatial scale from the local to the continental, from Easter Island to all of Africa,
with many stops in between. Every inhabited continent is represented, and most
of the great historic centres of human population. As a result, the book deals with
a wide array of cultures, political economies and landscapes. It is a step towards a
world environmental history of soils.
Lastly, this book is an interdisciplinary collaboration. Any approach to
the study of humankind and soils requires the methods and insights of multiple
scholarly disciplines. The arrangement of inquiry and knowledge into discrete
disciplines, which owes a little to Aristotle and lot to German universities of the
nineteenth century, has been an enormously rewarding strategy for at least 150
years. But it surely hides as much as it reveals. And no scholarly discipline is ad-
equately equipped (nor any single scholar as far as we know) to cope with the range
of methods, perspectives and data that is appropriate to the subject of the linkages
between humankind and soils through history.
The history of soils is perhaps the most neglected subject within environmental
history (McNeill and Winiwarter 2004). With this book we hope to address that
neglect. We also hope to provide a long-term perspective on soil-society relations
that may be helpful in putting soils more securely on the agendas of environmental
policy. Soil ecosystems lie, albeit uncharismatically, at the root of human existence.
But modern behaviour strongly suggests there is much as yet not understood about
soils and their links to society, and much that is understood that is imperfectly
practised. Soils are a very slowly renewable resource, and the use we make of them,
today as in much of the past, is often quite unsustainable. Understanding this in
all its dimensions, including its historical ones, is a step on the way to putting the
soilsociety relationship on a sounder basis.
Soils, Soil Knowledge, and Environmental History
ACKNOWLEDGEMENTS
This book is the outcome of a collaborative effort over several years. First, we would like to
thank all authors for their work in summarising and presenting knowledge from different
parts of the world in the context of this book.
We are very grateful to the Breuninger Foundation, Stuttgart (Germany) for their gen-
erosity in hosting a workshop on Wasan Island in August 2003. Most of the papers of this
volume were presented and discussed there. We would in particular like to thank Helga
Breuninger and her team for their hospitality and interest in soil matters.
It was a great pleasure to co-operate with Andrew and Alison Johnson of The White
Horse Press, who were instrumental in turning the idea of an interdisciplinary environmental
history of humansoil interactions on a worldwide scale into reality.
Verena Winiwarter acknowledges financial support for her work in contributing to and
editing of this volume from the Austrian Academy of Sciences, APART fellowship program,
Grant 10916, Precious Dirt Underfoot. A History of Soil Knowledge.
Production of maps for the Introduction and the chapters on Mesoamerica was made
possible by a grant from Georgetown Universitys Walsh School of Foreign Service.
REFERENCES
Blaikie, Piers and Harold Brookfield. 1985. Land Degradation and Society. London:
Routledge.
Braudel, Fernand. 1972. The Mediterranean and the Mediterranean World in the Age of Philip
II. New York: Harper & Row, 2 vols.
Gostomski, Anzelm. 1588 [1951]. Gospodarstwo. Wroclaw: Wydawn. Zakladu Narodowego
im. Ossolinskich, 1951.
Hillel, Daniel. 1991. Out of the Earth: Civilization and the Life of the Soil. Berkeley: Uni-
versity of California Press.
Hyams, Edward. 1952. Soil and Civilization. London: Thames and Hudson.
Lowdermilk, W.C. 1953. Conquest of the Land through Seven Thousand Years, Agricultural
Information Bulletin no. 99. Washington DC: US Department of Agriculture.
McNeill, J.R. and V. Winiwarter. 2004. Breaking the Sod: Humankind, History and Soil.
Science 304: 16279.
Neboit, Ren. 1983. Lhomme et lrosion. Clermont-Ferrand: Publications de la Facult des
Lettres de Clermont II.
Winiwarter, Verena and Winfried E.H. Blum. 2006. Souls and Soils: A Survey of World-
views. In Benno Warkentin et al. (eds), Footprints in the Soil: People and Ideas in Soil
History (Amsterdam: Elsevier), pp.10722.
7
An Introduction to Soil Nutrient Flows
Robert S. Shiel
ity deterioration invariably results in slower growth of plants and, by leaving the
soil surface exposed for longer, frequently increases soil erosion (McGregor et al.
1999), and this, by removing the nutrient-rich portion of the soil preferentially
(Stoltenburg and White 1953), has an even greater negative effect on fertility. In
the absence of modern fertilisers, declining fertility reduces crop productivity per
unit area and, if the supply of land is limited, can have serious consequences. A
shortage of nutrients leads to use of a narrow range of crops, which can be grown
under such conditions, over a large proportion of the total area. This narrowing of
the food base can result in catastrophic regional failures as happened in Western
Europe with potato crops in the middle nineteenth century ad (Burton 1948;
Bergman 1967). Where land is not in short supply, degradation of soil at one site
invariably leads to clearing of virgin land elsewhere, and this may, in its turn, suffer
the same fate; such a shifting cultivation system can only persist so long as there is
a fresh supply of unoccupied land (Nye and Greenland 1960).
The properties of the land in its virgin state depend on the material from
which it has formed (the parent material) and the extent of ageing to which it
has been exposed. The ageing is the result of chemical weathering of the soil
mineral particles and the loss of the potential plant nutrients, either by leaching
or by removal in the harvested crop. The extent of this weathering is one of the
fundamental factors used in determining the soil orders in the United States system
of soil classification (USDA 1999). The parent material from which the soil forms
can vary from siliceous sands through loessic or alluvial silts to nutrient-rich clays,
and depends very much on the underlying geology and processes of erosion and
deposition in the Pleistocene and Holocene (White 1997). The north and west of
Europe has been repeatedly affected by glacial episodes, the last of which finished
12,000 years ago, and which, to a variable extent, have replaced ancient, highly
weathered soils with glacial till (boulder clay), outwash sands or loess. While the
distribution of boulder clay largely follows the area glaciated, the loess and outwash
sands extend far to the south of the area covered by the ice sheets (Flint 1971).
Even in the peripheral areas to the south of the glaciation, the abrupt climatic
change, particularly in rainfall, at the end of the ice age caused widespread erosion
and the establishment of alluvial fans (Borsy 1990). Where the surface was not
disturbed by the ice sheets (Fitzpatrick 1963) then the residual soils are often very
acidic and low in plant nutrients and, like aged plants and animals, tend to have a
small productive potential. In these areas of very weathered and impoverished soils
it is in flood plains or where erosion has exposed fresh material that the soils are
likely to be most fertile. This is not to say that all of the young soils are fertile, for
those with a high content of silica sand contain few nutrients and evolve rapidly
to podzols (Paton et al. 1976) and have poor ability to retain nutrients or water,
in addition to being very acidic. Thus, Europe can be viewed as a patchwork of
soils, often with variation over short distances, between those youthful soils that
9
An Introduction to Soil Nutrient Flows
are naturally productive to those which require large amounts of inputs to achieve
a useful product. The proportion of these contrasting types will originally have
differed greatly between localities but this pattern has been affected by subsequent
exploitation of the land by farmers; their management may have improved condi-
tions in some, but in many other places has made the soil less productive.
An analogy with human ageing can help to appreciate the utility of soil. A
child has little productive ability and requires assistance, a juvenile tends to become
over enthusiastic, but an adult reaches a peak of productivity. As maturity progresses,
the range of work that can be done decreases, while in old age considerable assist-
ance is again needed. If we imagine an area of freshly exposed boulder clay at the
end of the ice age, this has a good content of mineral nutrients but lacks any system
of pores through which water, air and roots can move, and it contains little if any
of the essential nutrient nitrogen. The first plants to grow on such material have
to cope with a poor rooting environment and the lack of nitrogen the plants are
Stress Tolerators (Grime et al. 1988) and are likely to include nitrogen-fixing leg-
umes. The growth of these plants improves the conditions in the soil and the Stress
Tolerators are frequently replaced progressively by the more vigorous Competitor
plants that could not tolerate the original conditions. In the juvenile stage, the
conditions may be too rich for some species, but as the conditions become more
stable a maximum of productivity is reached. Throughout all this time nutrients
are being lost by leaching, and the soil progressively becomes more acidic and
depleted in nutrients such as potassium (White 1997). Plants which can tolerate
these oligotrophic conditions a different group of Stress Tolerators (Peet 1992)
tend to replace the Competitors, and growth again becomes restricted to a narrow
group of slower growing species. The period of time over which these changes occur
depends very much on the parent material, the climate and the topography. On
a level sandy site with a moderate rainfall, the whole sequence can be completed
in a few thousand years (Paton et al. 1976). As agriculture spread across Europe,
beginning from areas far to the south of those influenced by the direct effects of the
glacial episodes (Renfrew 1987), the young soils in most of the continent had gone
through the initial stages of formation, usually to the adult stage before they were
used. The main exceptions were in alluvial soils in river valleys and where erosion
occurred on hill slopes due to human activity. In these cases, the soil forming in
the alluvium contains much organic matter from the eroded soils and frequently
it stabilises rapidly into a fertile productive soil profile while the eroded hillsides
are rejuvenated sometimes to the infant stage!
Although crops and animals for food, fibre or fuel can be produced at all of
these soil development stages, some are better suited than others to each stage. If
species suited to an earlier stage or one which has not yet been reached is desired,
then the soil must be amended by the user to provide the appropriate conditions.
Examining first the selection of the most appropriate soil types, we have to con-
10
Robert S. Shiel
sider the crop species available. The cereals that are the backbone of modern food
production systems tend to grow best on relatively mature soils. If the nitrogen
supply is excessive, the plants grow too tall and fall over (lodge) making the crop
difficult to harvest and also prone to attack by predators and by fungal diseases
in the damp conditions near the soil surface. Modern wheat suffers less from this
problem because the varieties grown today are short and do not lodge so easily
this has allowed farmers to increase the nitrogen supply and hence the yield
(Percival 1974). On more mature soils that have become acidic then rye and oats
can be grown (MAFF 1979); there is also a form of barley, bere (Jarman 1996),
that is tolerant of soil acidity. Acidic soils, however, tend to have relatively poor
nitrogen (Alexander 1961) and phosphorus (Wild 1988) availability, so ensuring
an adequate nutrient supply under such conditions is not easy. Many plants grown
for fruit grow better in soils with a somewhat restricted nitrogen supply (MAFF
2000) strawberries for example will produce masses of foliage in fertile soil, but
the fruit is hard to find and tends to rot in the moist atmosphere created by the
rampant foliage. Most legumes prefer soil with a high pH and good supply of
available phosphorus but frequently compete poorly with more competitive spe-
cies if the nitrogen supply is good (Cooke 1975); the answer is therefore to grow
legumes after crops which have removed much nitrogen from the soil unless the
soil contains relatively little organic matter.
This in effect has introduced the concept of crop rotation as a means of
manipulating the nutrient supply to advantage the species that we desire. Rotation
will also reduce soil erosion (Stoltenburg and White 1953) and, by providing a
diverse range of crops, will reduce the risk of a catastrophic failure. While a rota-
tion has the objective of returning the soil properties at the end of the cycle to the
state that they were in at the beginning, and as such returns most of the nutrients
removed in animal feeds via the manure, the sale of crops for human consumption
in distant urban areas results in local loss. In the case of nitrogen these losses can be
made good by fixation from legumes within the rotation but other nutrients will be
depleted unless an external source is available. Recycling of urban wastes has only
ever benefited the areas close to the towns and is limited today by a variety of con-
cerns involving disease organisms and toxins in the wastes (Lampkin 1990). Other
management methods available to the farmer operate to modify the conditions in
the soil to render it more similar to those in which the species desired is most likely
to succeed; these techniques include drainage, liming, manuring, marling and, in the
extreme, warping (Russell 1915). The increased use of artificial fertilisers since the
Second World War has arguably solved the nutrient management problem, but by
removing the need for the rotation of crops and the production of animal manures
has exacerbated the other soil problems that are related to nutrient management.
Thus, we have more severe erosion, decreasing soil organic matter, and associated
with this poorer soil structure, increased drainage and soil management problems
11
An Introduction to Soil Nutrient Flows
(MAFF 1970), increased droughtiness and leaching of nitrates into water supplies
(Addiscott et al. 1991). To continue the analogy with human ageing used above,
liming, manuring and warping can be related to rejuvenation, rotation of use
between crops and grass can be related to taking exercise or a recreational change,
while the repeated growing of extractive crops, particularly with the use of large
amounts of fertiliser, could be related to self abuse!
Recently, there has been widespread and increasing interest in organic farm-
ing (Lampkin 1990) which emphasises the use of crop rotation and maintenance
of soil organic matter, and uses natural resources, such as limestone and phosphate
rock, to increase the nutrient content. Although this is argued by its proponents to
greatly improve the quality and sustainability of the soil, it is only returning the land
use system to one that was common in the early years of the twentieth century ad
(Fream 1919). It remains therefore to be seen how well the use of such rotations,
whether in the nineteenth or twenty-first centuries, can satisfactorily ameliorate
soils and maintain them in a sustainable condition while providing large amounts
of high quality food.
References
Addiscott, T.M., D.S. Powlson and A.P. Whitmore. 1991. Farming, Fertilizers and the Nitrate
Problem. Oxford: C.A.B. International.
Alexander, M. 1961. Introduction to Soil Microbiology. New York: Wiley.
Bergman, M. 1967. The Potato Blight in the Netherlands and its Social Consequences
(18451847). International Review of Social History, 12: 391431.
Borsy, Z. 1990. Evolution of the Alluvial Fans of the Alfold. In A.H. Tachocki and M.
Church (eds) Alluvial Fans: A Field Approach (Chichester: Wiley), pp. 22946.
Burton, W.G. 1948. The Potato. London: Chapman and Hall.
Cooke, G.W. 1967. The Control of Soil Fertility. London: Crosby Lockwood.
Cooke, G.W. 1975. Fertilizing for Maximum Yield. London: Granada.
Flint, R.F.1971. Glacial and Quaternary Geology. New York:Wiley.
Fitzpatrick, E.A. 1963. Deeply Weathered Rock in Scotland, Its Occurrence, Age, and
Contribution to the Soils. Journal of Soil Science, 14: 3343.
Fream, W. 1919. Elements of Agriculture, ed. J.R. Ainsworth-Davis. London: John Murray
(10th edn).
Grime, J.P., J.G. Hodgson and R. Hunt. 1988. Comparative Plant Ecology: A Functional
Approach to Common British Species. London:Unwin Hyman.
Jarman, R.J. 1996. Bere Barley: A Living Link with 8th Century? Plant Varieties and Seeds,
9: 1916.
Jenkinson, D.S. 1988. Soil Organic Matter and its Dynamics. In A. Wild (ed.), Russells Soil
Conditions and Plant Growth, 11th edn. (Harlow: Longman), pp. 564607.
Lampkin, N. 1990. Organic Farming. Ipswich: Farming Press.
12
Robert S. Shiel
Liebig, J. 1843. Chemistry in its Applications to Agriculture and Physiology, edited from the
manuscript of the author by Lyon Playfair, 3rd edn. London: Taylor and Walton.
MAFF (Ministry of Agriculture, Fisheries and Food). 1970. Modern Farming and the Soil.
London: HMSO.
MAFF. 1979. Fertiliser Recommendations. London: HMSO.
MAFF. 2000. Fertiliser Recommendations for Agricultural and Horticultural Crops (RB209).
Norwich: The Stationery Office.
McGregor, K.C., R.F. Cullum and C.K. Mutchler. 1999. Long Term Management Effects
on Runoff, Erosion, and Crop Production. Transactions of the ASAE, 42: 99105.
Nye, P.H. and D.J. Greenland. 1960. The Soil under Shifting Cultivation, Technical Com-
munication 51. Farnham Royal: Commonwealth Agricultural Bureau.
Paton, T.R., P.B. Mitchell, D. Adamson and R.A. Buchanan. 1976. Speed of Podzolisation.
Nature, 260: 6012.
Peet R.K. 1992. Community Structure and Ecosystem Function. In D.C. Glenn-Lewin,
R.K. Peet and T.T. Veblen (eds), Plant Succession (London: Chapman and Hall), pp.
10351.
Percival, John. 1974. The Wheat Plant:A Monograph. London:Duckworth.
Renfrew, C. 1987. Archaeology and Language:The Puzzle of Indo-European Origins.
London:Cape.
Russell, E.J. 1915. Soil Conditions and Plant Growth. London: Longmans, Green and Co.
Stoltenburg, N.L. and J.L. White. 1953. Selective Loss of Plant Nutrients by Erosion.
Proceedings of the American Society of Soil Science, 17: 40610.
Thirsk, J. 1997. Alternative Agriculture:A History From the Black Death to the Present Day.
Oxford:Oxford University Press,
USDA. 1999. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting
Soil Surveys, 2nd edn. Washington, DC:US Dept. of Agriculture, Natural Resources
Conservation Service.
Wild, A. 1988. Plant Nutrients in Soil: Phosphate. In: A. Wild (ed.), Russells Soil Condi-
tions and Plant Growth (Harlow: Longman), 11th edn, pp. 695742.
White, R.E. 1997. Soil Formation. In Principles and Practice of Soil Science (Oxford:
Blackwell), 3rd edn.
13
Exploitation and Conservation of Soil in South Asia
R.J. Wasson
Introduction
teenth century, during the time of Mughal rule, a rich literature written in Persian
focused on management of the rural economy largely for revenue. There is also
a treatise devoted to agriculture. During the British period (c. 17601947), the
relevant literature includes documents designed to steer the management of the
rural economy, again largely for revenue. Scientific papers in the newly developed
tradition of western science also appear, along with the observations of travellers
from various European countries.
Almost all of the pre-British literature refers to what is now Bangladesh,
northern India and northern Pakistan, with few accounts of central and southern
India. For continuity, only those documents from the British period that refer to
the same areas as those described in earlier documents have been used, with one
exception.
The degree to which this literature is normative or reflects agricultural practice
is not easy to determine. The Brahmanical tradition of stating what ought to be
suggests that at least some of the Sanskrit literature is normative. But this literature
also contains a lot of detail about methods for tilling soils, collecting water, land
capability, and also relationships between soil, land, water and vegetation. It is dif-
ficult to interpret this as anything but a record of common observations and practice
(cf. Elvin 1993 for China). Some of the literature provides methods for predicting
rainfall, and a calendar for agricultural activities that is astrologically based. Given
the significance of astrology in modern Hindu society, and its early beginnings, it
is likely that this literature reflects practice. There are however many passages that
are essentially spiritual, in which the farmer is urged to follow good ways.
From the thirteenth century onward a large proportion of South Asias popula-
tion was Muslim. The Muslim literature appears to reflect agricultural practice and
also to describe commonly used methods of soil classification and land capability
assessment. One exception is the only solely agricultural treatise of the seventeenth
century, which contains much of a scholarly nature that may have been the musings
of the writer rather than a reflection of widespread beliefs and practice.
The literature of European travellers and of the British East India Company
and British government bureaucracy reflects practice and classification, suitably
biased by the cultural lenses of the authors and their often commercial purposes.
Even the authors of the scientific literature serve their employers, and this material
must be read in the light of its purpose. Yet the British material is easier to interpret
than much that came before, simply because it is closer to our modern experience
and it is rich and abundant, providing many voices and views that can be cross-
checked. The audience for this literature is also clear.
For the local and in the main pre-European literature, the audience is not
clear. The earliest literature was almost certainly a codification of oral tradition.
The Vedas were, and still are, recited for religious purposes well after they had been
15
Exploitation and Conservation of Soil in South Asia
written down: a result of illiteracy and adherence to long tradition as illustrated by,
for example, the family name Trivedi for people who could recite three Vedas.
Were the agricultural treatises written for bureaucrats, literate farmers, or
for students? Were they committed to memory and recited by farmers? They may
have occupied in South Asia the same place in society as the agricultural almanacs of
Europe and North America, read by a few and disseminated by word of mouth.
The themes of this chapter are in summary: soils in agriculture, soil clas-
sification and land capability, soil degradation, and the overarching concepts of
exploitation and conservation. The narrow range of conservation issues is also
explored, and some insights provided by a comparison with the environmental
history of China. Some attention is also paid to human responses to soil changes,
mostly from the British period for which documentation exists. Finally, the modern
concept of a pre-British balanced agroecosystem is examined.
Before dealing with the written evidence, the chapter begins with a brief and
necessarily inconclusive summary of the relationship between hunter-gatherers and
soil. This leads into the evidence from early agriculture and the rise of urban centres
during the Harappan period. While the Harappans were literate, their writing has
not been deciphered. We must wait until the Vedas, starting around 1000 bc, for
the first decipherable written accounts.
The Evidence
Hunter-Gatherers
The place of soil in the worldview of hunter-gatherer societies is only known from
modern examples. In South Asia, hunter-gatherer peoples are now very rare, and take
part in cultivation and the market economy to varying degrees. There is therefore
almost nothing known of the place of soil in the pre-agricultural worldview of such
people. We therefore turn to archaeological records, almost solely from Pakistan,
northern India and Bangladesh (Figure 3.1).
The Harappans
The appearance in the Indus River valley of the Harappan Civilisation c. 3200 bc,
marked by extensive, well-planned cities (Kenoyer 2003), had precursors in
the beginnings of agriculture in Baluchistan (in what is now western Pakistan) c.
7000 bc (Possehl 1997). Shinde (2002) sees the south-west Asian agro-pastoral
system of wheat, barley, cattle, sheep and goats reaching Baluchistan by c. 7000
bc. Small villages appear with stone-banked fields, mud brick houses, and rituals
indicated by figurines and burial. Soil was clearly central to the agricultural life
of these people, at a time when climate was highly variable but becoming wetter
16
R.J. Wasson
(Singh et al. 1990). But we can only infer a system of cultivation, moisture reten-
tion and erosion control afforded by the terraced fields, and possibly fallowing.
There is nothing in the words of the people of their time.
The earliest villages appear along the Ravi River in the Indus valley c. 3300
bc, according to Kenoyer, (2003). These people herded cattle and cultivated wheat,
barley, legumes and sesame and are sometimes described as representing the Early
Harappan or Early Indus period of settlement (Possehl 2004). Between 2800 and
2600 bc the city of Harappa thrived economically and technologically. Exquisite
luxury goods such as beads and glazed pottery were produced, along with fine figu-
rines painted with what appear to be representations of woven fabric. Merchants
17
Exploitation and Conservation of Soil in South Asia
used seals of clay, and writing appeared. The determination of value by weight
developed, possibly for taxation or tribute.
Agriculture and grazing supported the economy of early Harappa, along the
Ravi River floodplain, but trade was clearly an important source of wealth based on
the high quality technological achievements of Harappan craftsmen. Granaries in
the larger cities evidence a surplus of agricultural production. Brick buildings and
the use of wood in construction show that forests were also being used, presum-
ably in the vicinity of the cities. From 3200 to 1900 bc, cities scattered over the
Indus Valley and to the east were nestled in a landscape of villages with cultivated
fields and grazing. The cities appear to have been abandoned c. 1900 bc, an event
described by many archaeologists and historians either as collapse or eclipse of
civilisation. Wheeler (1947) believed that a massacre had occurred at the great
Harappan city of Mohenjodaro, and that the Harappan civilisation had collapsed
as a result of invasion by the Aryans. He drew from the Rigveda, the oldest of the
Vedas written in Sanskrit with a largely spiritual and moral philosophical content.
Wheeler viewed the Rigveda as history, and saw in the destruction of ninety forts
by Indra, the Aryan warrior as puramdava (fort destroyer), documentary support
for an Aryan invasion.
Other explanations of the demise of the Harappan cities included climate
change or damming of the Indus by a fault (Raikes 1965) that resulted in cities
along the river being engulfed by mud. The latter is implausible (Wasson 1984).
Singh (1971) found evidence of climatic drying at about the time of abandonment
of the cities from pollen preserved in salt lakes in the Thar Desert. The science of
climate change in the region has subsequently been strengthened by Singh et al.
(1990) and Wasson et al. (1984). But the timing of the change has been questioned
by a detailed study of the geochemistry of Lake Lunkaransar, also in the Thar (Enzel
et al. 1999). Perhaps the most damaging problem for the argument that climate
change led to the Harappan collapse is that the climate during the period of great-
est vitality was not very different from that of today. Also, the cities were on rivers,
and the agriculture benefited from annual floods which continued even in a drier
climate because the rivers rise in the Himalayas.
Fairservis (1967) used estimates of the population size and calorific need of
the city of Mohenjodaro to calculate the area required for cultivation, the quantities
of fodder needed for cattle, and the amount of timber needed to fire the bricks for
the city and to keep its cooking fires alight. He concludes that the Harappan city
was in a precarious economic situation (which was) a significant reason for the
downfall of one of the worlds earliest civilisations. Too many demands were being
made on the soil and water resources of the area, and in addition the removal of
gallery forest (the trees along the river margins) would have exacerbated flooding,
according to Fairservis, and would have made agriculture difficult.
18
R.J. Wasson
Valley for settled agriculture can be compared with dates between c. 3500 and 2000
bp (2000 to 0 bc) in Southern India (Shinde 2002).
The evidence for a progressive shift of the Harappan culture from west to
east and then the continuing evolution of agriculture in the Ganges Valley is clear.
Urbanisation reappears in South Asia in the Ganges Valley c. 500 bc (Agrawal 2002;
Erdosy 1998) some 4500 years after settled agriculture developed in this region.
The mix of cultivation and pastoralism is also clear from the archaeological record
during the entire period of 8,500 years, as seen earlier. This land use pattern is also
reflected in the Vedas.
and the Vishnu Purana to the first century ad (Tharpar 1984). The Grihya Sutras
probably date from the second half of the first millennium bc (Thapar 1984), the
White Yajurveda to about 800 bc, the Brahmanas to ad 13001400, the Ramayana
to 400500 bc (Fosse 1997), and the Ashtadhyayi to the end of the fourth century
bc (Basham 1971). These texts therefore span about 2600 years.
It is generally accepted that the Rigveda is the oldest text at about 1000
bc, although this can never be known with certainty (Leach 1990). Nonetheless,
analysis of its vocabulary and content relative to other texts supports the view that
it is the oldest, even if the date is uncertain (Fosse 1997). Agriculture at this time
took account of soil type, as would be expected, and ideas had developed about
rivers and their floodplains. Half a millennium later the Ramayana records ideas
about catchment processes that affect water quality.
The earliest literature refers to both deserts and wetter country, names loca-
tions that some scholars believe can be identified, and provides some evidence of an
eastward migration of people. These pieces of evidence have been woven together
with archaeological finds to reconstruct a migration of agricultural people along
the Ganges Valley into the forests and swamps of Uttar Pradesh and Bihar (Thapar
1984; Sharma 1996). The internal consistency of the argument supports the idea
that the texts were written after the de-urbanisation of the Harappan civilisation,
although it is possible that some of the early literature draws upon the Harappan
oral tradition. It seems highly unlikely that any of the early literature describes
Aryan people who overran the Harappans.
A major social change occurred in the first millennium bc: states appeared
in the Ganges Valley. The change was the result of many factors (Thapar 1984).
The migration of people along the Ganges Valley provided readily available new
land, presumably because the people in the forest were few in number and unable
to resist the migrants. Therefore, conflict did not arise because the Vedic people
were not crowded, nor was there rapid demographic increase. The Vedic people
were not strongly socially stratified, and external conflict was limited, hence no
necessity for states existed. The rise of rice agriculture and irrigation probably
had some effect on social stratification, because landowners differed from labour-
ers. Internal tension caused by this stratification may have played a part in state
formation. When the head of a household (the grihapati) was transformed into a
landowner and participant in trade, and the shudra into a peasant cultivator and
artisan, stratification and tension increased. Shudras, the lowest of the four classes
(priest, brahmin; warrior, kshatriya; peasant, vaishya; and shudra) is a labourer who
is somehow subservient in the Rigveda (Thapar 1984).
Intensive use of the soil for agriculture that was partly dependent upon
irrigation in the first millennium bc contributed to social stratification embodied
in the varna system of classes. Growing wealth and the development of cities in
22
R.J. Wasson
the second half of the first millennium bc required defence. And so the states were
born, to reduce conflict within a socially stratified system and to defend cities.
Sharma (1996) adds an element to this account of state and varna formation:
iron. With the development of agriculture in the Ganges Valley, iron tools became
more common. Iron appears in the mid-Ganges plain c. 800 bc, and in the eastern
plain 1200600 bc. Sharma believes that iron was first used in war and hunting,
and only from c. 600 bc was iron used by farmers and artisans. The clay soils of
Magadha (south of the Ganges, east of the Son River, and west of the Hazar Tuki
River, bounded to the south by the Vindhyan Hills and Chotanagpur Plateau)
and the plains north of the Ganges in Bihar and eastern Uttar Pradesh were too
hard for wooden ploughs, it is claimed. The Manusmriti, a compilation of Hindu
law composed in its final form in the second or third century ad (Basham 1971)
advises against the use of iron-tipped ploughs because they injure the Earth and
its creatures. This appears to be reaction to a technology that was more intrusive
than others, and may reflect a forest dwellers view of nature.
Iron allowed the cultivation of larger areas, especially for wet rice, and there-
fore greater food production. The greater complexity of society, and particularly the
varna system, could thereby be maintained. The postulated increasingly intensive
use of the soil is therefore seen as a contributor to state and varna formation.
Makkan Lal (1986) counters Sharmas argument from a detailed study of
Kanpur District on the southern bank of the Ganges. Lal focused on the argument
that iron tools were needed to clear the thick monsoon forest of the mid-Ganges
Valley to allow cultivation. Lal sees no change in iron technology from the Painted
Grey Ware period (ending c. 500 bc) to the Northern Black Polished Ware (NBPW)
period (500160 bc) when agriculture spread, cities arose and states formed. Dur-
ing the NBPW period, most settlements were small with fewer than 500 people.
Many were along river banks where soils are coarsetextured and easy to plough,
and others were near lakes where forest was often absent. Most agriculture could
therefore be carried out without forest clearing. Lal concludes that urbanisation,
and by inference state and varna formation, was the result of social and economic
forces that generated an economic surplus. This idea is an echo of the traditional
view of the rise of cities in the Near East, and has none of the sophistication of
Thapars explanation involving many factors that produced social stratification,
tension, defence and the need for a state. As with the decline of Harappan cities,
the reasons for rise of states in the Gangetic Plain remains a mystery.
see the forests as sacred. These oral traditions, codified in the Manusmriti, along
with other ancient texts, continue to energise modern-day environmental activit-
ists (Prime 2002).
These proto-ecological ideas had no discernible impact on either the Hindu
or Islamic (Mughal) rulers who came later, as we will see.
Paddy was extensive in Bengal, Assam and north Bihar, with wells, ponds
and tanks supplying irrigation water. Sugar-cane was also widely grown in Bengal,
Bihar, Uttar Pradesh and further west. There are also records of wheat and barley
in Bengal.
Widespread paddy, irrigation systems and instructions in the Krishi com-
bine to suggest an intensive agricultural system in which soil preparation was
exacting.
Unfortunately the identity and purpose of Parashara, the texts author, are
unknown, although the first passage says that the Krishi is for the benefit of
farmers (Sadhale 1999). It is clear from the text that the author was an authority
on astrology, cattle husbandry and meteorology, as well as agriculture. Sadhale
claims that the Krishi is like a farmers almanac, containing information arranged
by season and month. But how many farmers could read the Krishi? The language
is simple, possibly enabling its memorisation and recitation. It is remarkable in
one particular aspect; it is solely concerned with rain-fed agriculture, suggesting
that it drew on experience in the better-watered regions, such as Bengal. But Nene
(in Sadhale 1999) suggests that it refers to northern Pakistan, on the basis of the
crops mentioned in the text.
Of the 243 verses in the Krishi, 117 have an astrological component (Nene
in Sadhale 1999). Sixty-nine relate to prediction of rainfall. Also, some 27 verses
deal with cattle, showing the importance of pastoralism in the rural economies.
Ploughing by cattle is to be strictly controlled by astrology, as are many other as-
pects of agriculture. Manuring by dried cattle dung is clearly described, and crops
will yield little without it. Other topics covered include water harvesting and water
retention by bunding, weeding, plant protection against disease, threshing, con-
struction of ploughs, and many festivals and ceremonies that should accompany
parts of the agricultural cycle.
return. A drab story of endemic warfare between rival dynasties (Basham 1971,
p. 71) followed, until the beginnings of incursions by Arabs and Turks starting in
ad 712. Between ad 1001 and 1027, Mahmud raided deep into northern India as
seen earlier. He annexed Sind and Punjab. The remainder of northern India was
independent, but divided between three chief kings who were at war: Prthvijara
Cahamana, Jayaccandra Gahadavala and Paramardideva Candella. After one failed
attempt in 1191, the Muslim invaders from Afghanistan overcame resistance as
far east as Bengal by 1203. Nepal, Assam, Kashmir and Orissa maintained their
independence, cut off from the Ganges Plain by natural boundaries. Until the
eighteenth century, Muslim rulers dominated northern India.
During this troubled time, the Kashyapiyakrisisukti was written. This is an
agricultural treatise, written mostly between ad 800 and 700 although there may
be later interpolations (Wojtilla 1985). The author, Kashyapa, was probably one
of a group of Brahmins of the Vaikhanasa school, followers of Lord Vishnu. Most
of the writings of the Vaikhanasa school were of a socio-religious kind and their
works were designed as a code of conduct, intended for other Brahmins, and for
government and village officials (Ayachit 2002).
The Kashyapiyakrisisukti is a detailed account of methods of cultivation of
a wide range of crops, both irrigated and unirrigated. While highly practical, the
text is set within a larger purpose. The production of grains and other vegetation is
said to be the sole purpose for (the) highest fulfilment of the Earth. It stresses
support for agriculture by the ruler, particularly identifying suitable land, building
reservoirs and planting trees on their banks, constructing canals, making seed avail-
able, donating land to weaker people, arranging markets, standardising weights
and measures, and afforestation. This list reminds us of the Arthashastra.
Once again the concepts of land capability appear. The importance of good
land is emphasised because it yields good results to everyone, confers good
health on the entire family, and causes growth of money, cattle, and grain. The
ruler should appoint knowledgeable people (interestingly from any caste) to as-
sess land for cropping. Good land should be: devoid of stones and bones; a pliant
(plastic) clay, very unctuous (greasy) with reddish and black hue, and glossy with
water; neither too deep nor too shallow; conducive to speedy seedling emergence;
easy to plough and cultivate; water absorbent, and replete with beneficial organisms
such as earthworms; and devoid of thorns and cow dung, thickly set and compact,
and heavy when it was lifted. This account reveals a combination of a physical and
biological concept of soil.
Verses I 51 to I 54 (in the enumeration of Ayachit 2003) provide methods
for examining soil, some of which would be familiar to a modern soil scientist. At
a point, a hole is dug and the effects of the digging observed repeatedly. The char-
acteristics of the soil of note are its colour and uniformity of colour, the drops
of water (probably its turbidity), taste, fluidity, and its stickiness.
28
R.J. Wasson
of land and crops was the basis of this system. Sher Shah believed that the state
should encourage agricultural productivity, and cultivators who cleared forest and
woodlands received exemptions or reduced taxes. Those who either increased the
cultivable area or sunk wells received similar relief, so there was an incentive for
increased exploitation. During Akbars reign (15561605), taxes were in the form
of produce, taking into account variations between regions in yields.
The land and soil classification used by the Mughals was superimposed onto
classifications that have their roots in the Vedas. In north Gorakhpur District, due
north of Varanasi on the Gangetic Plain, the East India Company employee Francis
Buchanan Hamilton described the native classification of land and soil, and land
uses, in the eighteenth century (summarised by Bhargava 1999) (Table 3.1).
The soil types are straightforward, showing a clear local correlation between soil
and crops. Dorus was used for everything except paddy, unless it occurred where
water collects. Muttear was used for all crops, while balua was used only for wheat,
30
R.J. Wasson
millet and arhar. Khadir bangar was used for barley, lentils, rice and peas, while
humwar bangar was used for wheat and sugar-cane. Chowrear bhat was useful for
sugar-cane and corn.
Many of the Districts farmers were highly mobile, cropping banjar for
up to 3 years before the fertility was reduced and they moved on. This pre-fallow
period was also the time when the lowest rents were charged by landowners, and
so there was an incentive for migration. In the most northerly part of the area, this
migratory agricultural system was not tamed by the Mughals, and only changed as
a result of the British Settlement (see Hill 1997, for Purnea District).
In the sedentary agricultural areas, soils and land use were similar to an
infield-outfield system (Bhargava 1999). Land within 400 or 500m of a village,
called gongyer or goind, was well manured and irrigated annually, producing two
crops a year. Beyond goind was madhyam, bich or meeano. This less managed land
produced only one crop a year. The most remote and lowest lands, or the outlying
fields, were called pallo, which received neither manure nor irrigation. Winter rice
and pulses were grown in pallo. Generally muttear occurred in goind, dorus in mad-
hyam or meeano, and balua occurred in pallo near rivers. While the pre-Mughal land
and soil classification matched land use to land type (and therefore capability), the
land classification used by the Mughals was based on production. The capability of
the land to produce was not explicitly surveyed, simply the outputs. The detailed
records upon which revenues were based come to us in their most complete form
in the Ain-i-Akbari compiled by Abul Fazl and completed in 1598 (Blochmann,
1872). Other documents from the late seventeenth and eighteenth centuries provide
further revenue data and other insights (Habib 1982; Moosvi 1987).
The Ain used a threeway classification of land: tilled (kishta) or cultivated
(mazrua); and waste (uftada) in the sub-categories of cultivable waste and unculti-
vable waste (Trivedi 1998). The crop yields are given for land under continuous or
near-continuous cultivation (polaj and parauti respectively), and for each crop three
classes of yield are recorded: high (gazida, ala), middle (miyana) and low (zubun).
It is not clear on what basis these three classes are distinguished (Moosvi 1987).
Trivedi (1998) calculated the area of forest at the time of the Ain for the
subas of Agra, Awadh, Delhi and Lahore (see Habib 1982 for locations). Near-
natural vegetation, in many cases forest, covered over 50 per cent of the provinces
of Agra and Delhi, 64 per cent in Awadh and 72 per cent in Lahore. The area of
cultivation ranged from 14 per cent to 28 per cent in these subas, and the area of
uncultivable waste ranged from 47 per cent in Agra to 89 per cent in Allahabad.
A reasonable fraction of the land designated uncultivable waste probably included
soils recognised at the time as being uncultivable or barren. There was therefore a
concept of land capability at this time.
31
Exploitation and Conservation of Soil in South Asia
Moosvi (1987) makes the important point that uncultivable land is not a
rigid concept. Fertility and capability for cultivation is only absolutely limiting on
rock or where salinity is extreme. As population increases, inferior land is cultivated,
assisted perhaps by irrigation. The area of cultivable waste therefore should increase
through time, which Moosvi demonstrates by comparison of the Ain data with
those from the British figures of 190910.
The social structure of power, seen by Elvin (1993) as the most important
factor in Chinese environmental history, was also a major control in Mughal times.
The jagirdars were almost completely controlled by the Emperor, including the
methods of assessment and collection of land revenue (Habib 1963). The creation
and collection of surplus created Mughal wealth, and the demand for revenue
increased with time. Vast wealth and grinding poverty sat side by side as a result,
in the words of the Dutchman Pelsaert (quoted by Habib 1963).
Rotation of jagir every 3 to 4 years meant that long-term planning by jag-
irdars was impossible, and their interests were best served by extracting maximum
revenue as fast as possible. The Frenchman Bernier observed (also quoted by Habib)
of jagirdar behaviour: Let me draw from the soil all the money we can though
the peasant should starve or abscond and we should leave it, when commanded to
quit, a dreary wilderness.
The testimony of Pelsaert and Bernier is supported by other European observ-
ers of the seventeenth century, and by local writers. So great were the depredations
of at least some of the jagirdars that whole villages were deserted, land lay unculti-
vated because the oxen had been sold to pay jagir, and one observer reported that
many areas have become forests infested by tigers and lions (Habib 1963, p.
325). During this period there was some extension of cultivation into the tarai (the
swampy area on the Gangetic Plain near the Himalayas) and the delta in Bengal.
This increase of cultivation may have offset the area of abandonment, and so the
cultivated area in Aurengzebs time exceeded that in the Ain despite abandonment.
The relentless Mughal quest for revenue inspired soil abuse that, in turn, required
that new soils be brought into use.
While the original sources need to be further consulted, we can infer that
soils over large areas in the seventeenth century were not cultivated or manured,
and some were re-vegetated presumably by thorny scrub of the kind that can now
be found on poorly managed land. The migration of peasants, to escape unrea-
sonable revenue demands, led to the farming of new areas. A peasant moving to a
new area might even obtain assistance to cultivate, a practice eventually outlawed
(Habib 1963, p. 329).
Some areas of uncultivated soil therefore were cultivated not as a direct result
of Mughal concessions but as a result of oppression. Areas previously cultivated were
abandoned, sometimes to be re-occupied. But the extent of areas involved and the
impacts on soils are not explicitly known (Raychaudhuri and Habib 1982).
32
R.J. Wasson
Saltpetre
Soil was not only the basis of agriculture in Mughal India, but also supplied
another product: saltpetre. This is a naturally occurring chemical compound of
potassium and nitrate (KNO3), although sometimes the name saltpetre is applied
to sodium nitrate (NaNO3). Indian saltpetre is mainly the potassium compound
(Sarkar 1975).
The Ain records saltpetre mines along the Ganges and to its north scattered
about on the alluvial plains of rivers in Bihar (Habib 1982). It also occurred south
of the Ganges. Other major sources were north of Agra on the Doab, and north
of Ahmedabad in Gujarat.
33
Exploitation and Conservation of Soil in South Asia
Saltpetre was used by the Mughal elite as the oxidising component of gun-
powder, for cooling water, and as fertiliser (Akbar 2000; Sarkar 1975). When the
British and Dutch began to trade in the Ganges Valley in the early seventeenth
century, saltpetre was one of the goods of interest to them. The Dutchman Pelsaert
and the Englishman Peter Mundy have left records of the source and method of
production of saltpetre. It was prepared from three types of earth: black, yellow
and white. But the best saltpetre was said to come from black earth, being free of
salt or brackishness (Sarkar 1975).
The entire focus of the Mughals was revenue and exploitation, so soils
were considered only as a contributor to the yield of different kinds of land. The
judgement of barren, particularly saline, lands as unproductive is also a reflection
of the focus on revenue. Saltpetre production from soil had a commercial and
military purpose.
34
R.J. Wasson
Sri Lanka
A brief account of agricultural development in the northern dry zone lowlands of
Sri Lanka is now warranted, because here we see the same features encountered
in northern India. The dry zone is the cradle of Sinhalese Civilisation, where the
development of tank irrigation touched almost every river valley (Ismail 1995).
From the fifth century bc to about the first century ad, settlers spread up the main
river valleys such as the Malwatlu-oya and the Mahaweli Ganga. Tanks allowed
the population to occupy all of the dry zone by the end of this period, coincident
with the spread of Buddhism.
Even more tank building occurred between the second and the fifth century
ad during a period free from foreign invasion. The first of the great tank-building
Kings, Vasabha (ad 65109), appears in the contemporary source the Mahavamsa
(the Chronicles, Geiger 1934). The author(s) of the Chronicles attribute to Vasabha
an objective for his tank and canal constriction: to make the land more fruitful.
These words are a gentler version of the motivations of the Mughals and the author
of the Arthashastra, but there is a common purpose.
But there appears to have been a further purpose, that of developing the ability
to repulse foreign invaders particularly from what is now India. In the Culavamsa,
the later part of the Mahavamsa, we learn of the King Sena II (ad 851885) that
From this time onward he made the Island hard to subdue by the foe and made
it increase in wealth like land of the Uttarakurus (Geiger 1953). The Uttarakurus
were a mythical people who lived in a utopian land (Ismail 1995). So the ability
to marshal armies, fund their activities and feed them, depended upon a strong
agricultural economy that in the dry zone required extensive irrigation. All of this
in the presence of Buddhism.
this worldview arose and soil science, geology and geomorphology developed in
Western Europe.
James Hutton, the son of a respected Edinburgh merchant, had his Theory
of the Earth published in 1795. The idea that soil was formed by the rotting of rock
first appeared in this work. This was a substantive challenge to the prevailing view
that the Earth was created for man, and so the idea that it could rot, and erode,
was untenable to the faithful. The so-called denudation dilemma was solved by
Hutton by supposing that erosion and soil formation were in balance, so erosion
did not remove all soil. The habitability of the Earth was thereby maintained
(Davies 1968).
Soil science had its origins in agricultural studies and in earth science and
chemistry. In the late nineteenth century, Russian scientists established many of
the principles by which soils are still explained. The idea of soil forming processes,
and soil zonalism in which soil is formed over time by the action of climate and
organisms on parent material (rocks and sediments) conditioned by relief were
firmly established in the USA by 1938 (Paton et al. 1995).
Prior to the first scientific accounts in the early to mid-nineteenth century,
the only written accounts from South Asia in which soil plays a role are those
of European travellers, missionaries and East India Company men. Early travel
accounts say next to nothing about soils (Wessels 1924; Archer 1980) One very
informative account does however exist from the eighteenth century. A keen observer,
mapmaker and Surveyor-General of Bengal was the Englishman James Rennell
(17421839). In a paper of 1781, Rennell provided a thoroughly modern account
of the behaviour of sediment in meandering and straight rivers. His observations of
the Ganges and Burrampooter (Brahmaputra) rivers are mostly morphological and
sedimentological. He notes how the rivers are enriching of adjacent lands
(p 89); presumably a reference to fresh sedimentation on floodplains. He recounts
how a meandering river removes sediment from its outer near-vertical bank and
deposits sediment on inner shelving banks. He describes how banks more than 30
feet (~9m) high in the dry season are swept away during floods, especially where
the soil is loose, causing a change in the course of the river. Rennell lived in Ben-
gal for eleven years, during which time he observed changes to both the Jellinghy
River and the Ganges. The latter shifted one bank the breadth of an English mile
and a half (~2.4km) in nine years. This is an astonishing rate of bank erosion of
~270m/yr, although Rennell notes that this is an extreme rate and the more usual
is between ~130 and 160m/yr. Rennell goes on to give an account of water flow
in a river bend, and of the origin of channel bifurcation. The meandering habit of
the Ganges is attributed to the looseness of the soil.
In his account of the delta of the Ganges-Brahmaputra, Rennell describes
the turbidity plume in the Bay of Bengal, produced by fine-grained sediments in
suspension in water, which extended a distance of twenty leagues (roughly 100km)
36
R.J. Wasson
into the sea. The deposition of mud and organic matter and sand produces banks
that are just below the surface of the sea, and one day will rise above the sea to be
cultivated. Rennells account also includes a plausible physical explanation of river
channel islands in the Ganges, which after only a few years accumulate enough
mould to be cultivable; that is, deposited mud and organic matter that provides
a natural fertiliser for annual crops. This system of opportunistic agriculture is
recorded in the Vedas, and in the Krishi-Parashara, and continues to the present in
the flooded or diara lands along most rivers of northern India. Today, cow manure is
used as fertiliser, and the people who cultivate this rudimentary soil have little else to
depend upon, unlike diara cultivators of earlier times who probably had additional
land on floodplains and elsewhere, given that the pressure on land was lower.
Rennell, writing at the same time as James Hutton, was formulating his
geologic and geomorphic ideas, which combined keen observation and an under-
standing of fluid mechanics and sedimentology to account for the behaviour of the
Ganges River in its lower reaches. Stratification of sediment exposed in eroded river
banks and in the delta is accounted for by appealing to differences in the respective
gravities of sand and finer particles producing strata of different texture depend-
ing upon the stage of flood. He provides an estimate of the suspended sediment
concentration in the Ganges at its height, wherein a glass of water contains one
part in four of mud. The concentration would have been less than 25 per cent of
course, because the mud contains water. Finally, in describing the hydrology of the
Ganges, Rennell draws upon the observations by Count Buffon to explain levees
on the Ganges as the result of precipitation of mud as the water purifies itself as
it flows over the plain.
Rennells paper, communicated to the Royal Society of London by the great
botanist Joseph Banks, must rank as one of the earliest scientific accounts of a river
and its sediments and soils anywhere in the world.
Other writers were more interested in the political and economic aspects of
India. Francis Buchanan Hamilton was one of the most travelled and productive
writers who worked for the East India Company (Allen 2002). His journey into
Nepal in 180203, from Bihar to Kathmandu, resulted in an account of the land-
scape, people, agriculture and politics of the mountain kingdom (Hamilton, 1819).
He described the rocks, using whatever knowledge he had of the young science of
geology, and listed the rock types found in the stream gravels. He described forests,
cultivated areas, and the oddities of soil such as a singular black ferruginous earth
on the banks of the Kosi River that was eaten by elephants and used by the local
people for ink. But we learn little else, except that cultivation is restricted to valley
floors where soils are fertile, water is available, and gradient workable.
Buchanan, as he was known, also wrote extensively of Bihar as it was in
181112. But once again the usual accounts are given of soil fertility variations,
37
Exploitation and Conservation of Soil in South Asia
crops and irrigation systems without the scientific analysis that was developing in the
ranks of British residents in India. But Buchanans purpose was to obtain informa-
tion relevant to the Company, and this was mostly political and commercial.
About 1820, Major General Sir Alexander Walker, originally an employee
of the East India Company, authored a detailed account of agriculture in various
parts of India (Walker 1997). Soils appear in this account in a number of ways.
Attempts to introduce English ploughs and other agricultural implements to the
Malabar area failed because the Mahrattah (Maratha) people found them to be
too heavy, thereby fatiguing their oxen. The soils were too dense for these ploughs,
and Walker celebrates the different kinds of ploughs, both drill and common, that
Indian cultivators use for different sorts of seeds and soils. The English generally
blamed the Mahrattahs prejudices, sloth and obstinacy for the failure. But Walker
saw the matter differently. The Indians already had a workable and adaptable system
for cultivation. They did not need English implements.
Walker described the native soil classification of Malabar, noting that soil
fertility without water is of little value, especially in a land where for half the year
the soil is hard and cohesive. In Malabar, soil is classified into three sorts:
Pasheemah Koor the highest quality soil consisting of rich clay. The quality
of the soil is discovered by digging a hole a yard (~1m) deep and wide. If
the soil is the right kind, then not all of the excavated soil will be needed
to fill the hole.
Rashee Pasheemah Koor this is of medium quality and will exactly fill the
hole. This soil is not as adhesive as that of the highest quality, the word
rashee meaning a mixture of earth and sand; and today it would probably
be called loam.
Rashee Koor the poorest soil that will not fill the hole, consisting of loose
sand.
Walker was astounded to discover that this method of determining quality should
accord with methods and ideas recommended by Europeans such as Lord Kairns
and Sir H. Davy, despite his otherwise favourable views of Indian agriculturists.
The same technique was used in the Roman Mediterranean (See Chapter 7)
There is one further reference by Walker to soil fertility from Bengal. Quot-
ing a Mr Colebrooke, Walker observes that the extraordinary fertility of the soil of
Bengal is a disincentive to anything but subsistence agriculture because of the ease
of producing crops. By contrast, in Scotland nature provided little benefit to the
farmer and as a result some of the finest agricultural improvements have occurred.
Walker does not tell us if such improvements have occurred in those parts of India
where nature has not been as kind as in Bengal.
38
R.J. Wasson
ology and soil properties were not explored, and still have not been in systematic
ways. Science has developed in a disconnected way. It was only in the mid-1960s
that a need for integrated research involving soil physics, chemistry, microbiology,
pedology, land use planning, soil conservation, soil management and soil testing
was perceived by land managers. The first nationwide survey of soil degradation
was published in 1994 by Sehgal and Abrol (eg. Velayuthum et al. 2002).
The separation between the science of soil for agriculture and for under-
standing soil erosion and sedimentation remains, although foresters working in
the Himalayas and a handful of scientists concerned with catchment management
have come closest to making the links.
A Balanced Agroecosystem
Gadgil (2001) and Prime (2002) among others have pictured the pre-British vil-
lage-based agroecosystem as one in balance. Prime quotes extensively from Banwari,
editor of Jansata, a Hindi daily newspaper published in Delhi, to support his case.
Banwari also published a book on forest culture (1992), in which he depicts the
Hindu idea of the world as a forest, drawing on the Ramayana and Mahabharata
epics that are rich in descriptions of the forests of the first millennium bc. The
intrusion of agriculture broke up the natural forest cover, so that gradually a three-
way classification of forest emerged: tapovan, where sages seek truth; shrivan which
provides prosperity; and mahavan, or the great natural forest where all species
find shelter. Once again we find resonances with the three pillars of sustainability;
economy, society (dominantly spirituality) and environment. The preservation of
these forest types makes a village a full entity, married to the idea of the panchavati,
symbolising earth, water, fire, air and ether (Banwari 1992).
Gadgil, one of Indias most insightful modern ecologists, provides less detail
than Prime and Banwari, and essentially no scientific evidence for the postulated
balanced agroecosystems. Examples of such scientific evidence comes from the
Himalayas in the hands of P.S. Ramakrishnan (pers.comm) and K.G. Saxena (pers.
comm), but nothing is known from the plains perhaps because these pre-modern
agroecosystems no longer exist to be studied. Bhargava (1999) sees no evidence
for equilibrium in the pre-British village agroecosystem. Peasants could see no end
to the natural vegetation, and were not natural conservators of the environment,
according to Bhargava. And the Mughal agricultural system scarcely favoured bal-
anced agroecosystems.
The descriptions of pre-British agroecosystems of the Gangetic Plain, and
the accounts by old villagers that this author has been privileged to hear, certainly
indicate a much greater forest cover, local water harvesting systems of high water
quality, a nutrient recycling system that maintained soil fertility and limited prob-
lems of salinity and sodicity prior to the advance of canal irrigation, waterlogging
and salinisation.
42
R.J. Wasson
Growing population
increased flooding
Summary
For most of the 3000 years of documentary records of northern India, soils have
been seen as part of the production system. The focus has been on simple classifi-
cation of soils for cropping and pastoralism, and for revenue calculation. At times
the links between soil and water quality, sedimentation and river change have been
noted but these links were secondary to the view of soil as a sustaining resource for
agriculture. At no time prior to the modern era does the role of soil in ecosystems
receive much attention, and it only appears during the British period as part of a
forest production system or as a buffer to both runoff and flooding. Environmental
problems such as reh and soil impoverishment receive little attention in the surviving
sources, except as a brake on crop yield, until the post-Independence period.
The idea of a pre-British equilibrium agroecosystem on the plains of northern
India is part of the nationalist reaction to the colonial legacy of the British. The lost
agro-ecosystems are items of reverence to be recreated. The slight available evidence
suggests that, if there was an equilibrium, it existed for 2000 years from well before
the Mughals to nearly the present day. As gauged by erosion rate estimates, it has
been the post-Independence drive for food-self-sufficiency that has dramatically
upset the putative equilibrium, not the pressure of British rule on agriculture.
Conservation for most of the period was focused on soil fertility and forest
resources. This is not surprising, given that most of the literature describes condi-
tions on the plains where these issues, along with water supply, were of paramount
importance. The geographically wide-ranging Kashyapiyakrisisukti mentions a role
for the ruler to reafforest mountain slopes, the only pre-British reference to this
idea. While it is not clear from the brief account by Kashyapa why such afforesta-
tion is desirable, it is possible that his experience (or accounts from others) of the
mountainous lands adjacent to the Indo-Gangetic Plains, and in central India, where
hills rise abruptly from plains, was like that of China. Elvin (1993) documents
many cases in eastern China, where hills rise abruptly from plains, of deforesta-
tion leading to massive erosion and flooding. The need to protect adjacent areas
of cultivation appears to have driven a call for conservation of the hill-slopes. If
the literature of South Asia were not dominated by the plains, similar reflections
might have occurred prior to the British period when interest in the commercial
opportunities of the Himalayas, in particular, drew their attention to the erosion
and flooding hazards of over-clearing.
While it has been shown that much of what passed for science in the de-
bate over Himalayan forestry was driven by the imperatives of bureaucratic power,
nevertheless many of the British were genuinely concerned about the impact on
the cultivated plains of land use in the mountains. This concern, and the debate,
continues to the present. Here, at least, we get a more complete view of the response
to soil changes, and of the many factors that contribute to the response. In the
earlier literature, the response of cultivators to the impost of the Mughal revenue
46
R.J. Wasson
collectors is visible, with apparently large-scale migration as the major response and
recolonisation of once cultivated soils by native vegetation and animals.
Manuring is the most pervasive response to soil fertility decline, with every
text describing techniques and sources of manure. That manuring is an early response
is clear from its appearance in the Vedas.
Exploitation of soil is deep in Indian culture and society, with the Kashyapi-
yakrisisukti of some 1300 years ago setting the tone. The earth is fulfilled to the
highest degree by the production of grains and other vegetation for human use.
It is therefore no surprise that India is struggling to include conservation issues in
the thinking and actions of its rural population, that are broader than those that
directly sustain life in agroecosystems.
This chapter has depended upon existing translations of works in Sanskrit
and Persian for the earlier literature. Most of the translators have been concerned
with social, economic and religious issues, and a re-reading of these works in the
original languages may draw out fresh insights into natural phenomena, includ-
ing soil, and the human relationship to them. This might best be done by expert
linguists working with soil scientists, geomorphologists, ecologists and historians.
Field surveys of current agricultural practice, enlightened by such a re-reading,
may allow comparisons with ancient practices and views of soil, showing where
traditional views still exist and where they have been supplanted.
The major new area of research is of the kind begun at Misa Tal. The sedi-
mentary record in lakes, rivers, tanks and reservoirs is a rich archive containing
information that can be compared with the documentary record. Calibration of the
pollen record in modern agroecosystems, such as has been pioneered by European
scientists, would permit a more quantitative reconstruction of land cover. From this
the state and treatment of soils could be inferred. Here is another multi-disciplinary
research field, involving palaeoecologists, earth scientists, agricultural scientists,
historians and linguists.
Such new research could not only aid our understanding of how people
have related to soil, and how this relationship has and has not evolved, but also how
the practices and attitudes of the past inform current opinion, and how present
practice might be improved.
AcknowledgementS
Dr. K Proust and Dr. Y.L. Nene for critical comments, McComas Taylor for checking the
orthography, and Margo Davies, Melinda Wall and Joanne Pinter for assistance with the
manuscript. Merrilyn Wasson for teaching me so much about India.
47
Exploitation and Conservation of Soil in South Asia
References
Agarwal, A.K., M.H. Rizvi, I.B. Singh, A. Kumar and S. Chandra. 1992. Carbonate Deposits
in Ganga Plain. In I.B. Singh (ed.), Gangetic Plain: Terra Incognita (Geology Dept.,
Lucknow University), pp. 3544.
Agrawal, D. P. 2002. The Earliest Agriculture and Pottery in South Asia. In Y. Yasuda (ed.),
The Origins of Pottery and Agriculture (New Delhi: Roli Books).
Akbar, R. (trans.) 2000. Nuskha Dar Fanni-Falahat (The Art of Agriculture). Agri-History
Bulletin, 3 (Secunderbad, India: Asian Agri-History Foundation).
Allen, C. 2002. The Buddha and the Sahibs. London: John Murray.
Archer, M. 1980. Early Views of India: The Picturesque Journeys of Thomas and William Daniell,
17861794: The Complete Aquatints. London: Thames and Hudson.
Ayachit, S.M. (trans.) 2002. Kashyapiyakrishisukti (A Treatise on Agriculture by Kashyapa).
Agri-History Bulletin, 4 (Secunderbad, India: Asian Agri-History Foundation).
Banwari. 1992. Pancavati: Indian Approach to Environment. Delhi: Shri Vinayaka Publica-
tions.
Basham, A.L. 1971. The Wonder That Was India: A Survey of the Culture of the Indian Sub-
Continent before the Coming of the Muslims (includes information on Sanskrit literature
in Language and Literature). New York: Hawthorne.
Bhargava, M. 1999. State, Society and Ecology: Gorakhpur in Transition 17501830. New
Delhi: Manohar.
Blochmann, H. (ed.) 1872. Ain-i Akbari. Calcutta, 2 vols. Osnabruck, Biblio Verlag,
1985.
Buchanan, F. 1936. An Account of the Districts of Bihar and Patna in 18111812. Patna:
Bihar and Orissa Research Society.
Chauhan, M.S., A. Khandelwal, S.K. Bera and H.P. Gupta. 1991. Palynology of Kathauta
Tal, Chinhat, Lucknow. Geophytology, 21: 1914.
Centre for Science and Environment. 1991. Floods, Flood Plains and Environmental Myths.
New Delhi: Centre for Science and Environment (State of Indias Environment: A Citi-
zens Report, no 3).
Choudhary, A.K. 1971. Early Medieval Village in North-Eastern India (ad 6001200).
Calcutta: Punthi Pustak, Patna University.
Davies, G.L. 1968. The Earth in Decay. London: MacDonald Technical & Scientific.
Eckholm, E.P. 1975. The Deterioration of Mountain Environments. Science, 189:
76470.
Elvin, M. (1993). Three Thousand Years of Unsustainable Growth: Chinas Environment
from Archaic Times to the Present. East Asia History, 6: 746.
Erdosy, G. 1998. The Prelude to Urbanization: Ethnicity and the Rise of Late Vedic Chief-
doms. In F.R. Allchin (ed.), The Archaeology of Early Historic South Asia: The Emergence
of Cities and States (Cambridge: Cambridge University Press), pp. 7598.
48
R.J. Wasson
Enzel, Y., L.L. Ely, S. Mishra, R. Ramesh, R. Amit, B. Lazar, S.N. Rajaguru, V.R. Baker
and A. Sandler. 1999. High-Resolution Holocene Environmental Changes in the Thar
Desert, Northwestern India. Science, 284: 1258.
Fairservis, W.A. 1967. The Origin, Character and Decline of an Early Civilization. Novis-
tates, 2302: 148.
Fisher, R.J. 1990. The Himalayan Dilemma: Finding the Human Face. Pacific Viewpoint,
31(1): 6976.
Fosse, L.M. 1997. The Crux of Chronology in Sanskrit Literature: Statistics and Indology, a
Study of Method, Acta Humaniora 21. Oslo: Scandinavian University Press.
Gadgil, M. 2001. Ecological Journeys. The Science and Politics of Conservation in India. New
Delhi: Permanent Black.
Geiger, W. (trans.) 1934. Mahavamsa or the Great Chronicle. London.
Geiger, W. (trans.) 1953. Culavamsa, being the more recent part of the Mahavamsa. Co-
lombo
Gupta, H.P. 1976. Holocene Palynology from Meander Lake in the Ganga Valley, District
Pratapgarh, U.P. Palaeobotanist, 25: 10919.
Habib, I. 1963. The Agrarian System of Mughal India (15561707). Bombay: Asia Publish-
ing House.
Habib, I. 1982. An Atlas of the Mughal Empire. Delhi: Oxford University Press.
Hamilton, F.B. 1819. An Account of the Kingdom of Nepal, Bibliotheca Himalayica, Series
1, Vol. 10. New Delhi: Manjusri, 1971.
Hill, C.V. 1997. River of Sorrow. Association for Asian Studies, The University of Michi-
gan.
Hutton, J. 1795. Theory of the Earth. Edinburgh: Scottish Academic Press.
Islam, M.M. 1997. Irrigation Agriculture and the Raj: Punjab, 18871947. New Delhi:
Manohar.
Ismail, M. 1995. Early Settlements in Northern Sri Lanka. New Delhi: Navrang.
Ives, J. D. and B. Messerli. 1989. The Himalayan Dilemma: Reconciling Development and
Conservation. London: Routledge/United Nations University.
Kangle, R.P. 1972. The Kautlya Arthasastra: An English Translation with Critical and Explanatory
Notes, Volume 2. Bombay University Press.
Kanwar, S.S. 1997. A Soil Scientist Reminisces: Fifty Years of Soil Science Research to Serve
Indian Agriculture. Asian Agri-History Journal, 1(2): 13552.
Kenoyer, J.M. 2003. Uncovering the Keys to the Lost Indus Cities. Scientific American,
289(1): 5867.
Lal, M. 1986. Iron Tools, Forest Clearance and Urbanisation in the Gangetic Plains. Man
and Environment, 10: 8390.
Leach, E. 1990. Aryan Invasion over Four Millennia. In. E. Ohnuki-Tiernery (ed.), Culture
Through Time: Anthropological Approaches. Stanford: Stanford University Press.
Moosvi, S. 1987. The Economy of the Mughal Empire, c.1595: A Statistical Study. Delhi:
Oxford University Press.
49
Exploitation and Conservation of Soil in South Asia
Nene, Y.L. and N. Sadhale. 1997. Agriculture and Biology in Rigveda. Asian Agri-History,
1(3): 17790.
Paton, T.R., G.S. Humphreys and P.B. Mitchell. 1995. Soils: A New Global View. London:
UCL Press.
Possehl, G.L. 1997. The Transformation of the Indus Civilisation. Journal of World Prehis-
tory, 11(4): 42572.
Possehl, G.L. 2004. India, Ancient-Indus Valley. In S. Krech, J.R. McNeill and C. Merchant
(eds), Encyclopedia of World Environmental History, Vol. 2 (FN) (New York: Routledge),
pp. 6725.
Prime, R. 2002. Vedic Ecology. Novato, CA: Mandala Publishing.
Raikes, R.L. 1965. The Mohenjo-daro Floods. Antiquity, 39: 196203.
Raychaudhuri, T. and I. Habib. 1982. The Cambridge Economic History of India, Vol. 1:
c.12001750. Cambridge: Cambridge University Press.
Rennell, J. 1781. An Account of the Ganges and Burrampooter Rivers. Roy. Soc. London
Phil. Trans., 71: 87114.
Richards, J. 1993. The Mughal Empire. Cambridge: Cambridge University Press.
Saberwal, V. 1999. Pastoral Politics. Delhi: Oxford University Press.
Sadhale, N (trans.) 1999. Krishi-Parashara (Agriculture by Parashara). Agri-History Bulletin,
2 (Secunderbad, India: Asian Agri-History Foundation).
Sarkar, J. N. 1975. Studies in Economic Life in Mughal, India. Delhi: Oriental Publishers.
Saxena, D. P. 1976. Regional Geography of Vedic India. Kanpur: Grantham.
Sharma, A. 1983. Studies in Alberunis India. Wiesbaden: Otto Harrassowitz.
Sharma, C., M.S. Chauhan, S. Sharma, M. Sharma and I.B. Singh. 2001. Proxy Records
of Holocene Vegetation and Climate Change from Sanai Tal, Central Ganga Plain,
Uttar Pradesh. Nat. Symp. Role of Earth Science Integrated Development and Related
Societal Issues. Geol. Surv. India. Spec. Publ., 65(111): 199202.
Sharma, R.S. 1996. The State and Varna Formation in the Mid-Ganga Plains. Delhi: Mano-
har.
Shinde, V. 2002. The Emergence, Development and Spread of Agricultural Communities
in South Asia. In Y-Yasude (ed.), The Origins of Pottery and Agriculture (New Delhi:
Roli Books), pp. 89118.
Singh, G. 1971. The Indus Valley Culture Seen in the Context of Post-Glacial Climatic
and Ecological Studies in Northwest India. Archaeology and Physical Anthropology in
Oceania. 6(2): 17789.
Singh, G., R.J. Wasson and D.P. Agrawal. 1990. Vegetation and Seasonal Climatic Changes
since the Last Full Glacial in the Thar Desert, Northwestern India. Review of Palaeobotany
and Palynology, 64: 35188.
Srivastava, P., I.B. Singh, S. Sharma, V. Shukla and A.K. Singhvi. 2003. Late Pleistocene-
Holocene Hydrologic Change in the Interfluve Areas of the Central Ganga Plains, India.
Geomorphology, 54: 27992.
Starkel, L., and S. Basu (eds). 2000. Rains, Landslides and Floods in the Darjiling Himalayas.
New Delhi: Indian National Science Academy.
50
R.J. Wasson
Thapar, R. 1984. From Lineage to State: Social Formations of the Mid-First Millennium bc in
the Ganges Valley. Bombay: Oxford University Press.
Thompson, M., M. Warburton and T. Hately. 1986. Uncertainty on a Himalayan Scale.
Milton Ash Edition.
Tripathi, T.M. 1990. Hydrology in Ancient India. Rookee: National Institution of Hydrol-
ogy.
Trivedi, K.K. 1998. Estimating Forest, Wastes and Fields, c. 1600. Studies in History,
14(2): 30111.
Vasu, S. C. (ed.) 1962. Paninis Astadhyayi. Delhi.
Velayutham, M., D.K. Pal, T. Bhattacharya and P. Srivastava. 2002. Soils of the Indo-
Gangetic Plains, India: Historical Perspective. In Y.P. Abrol, S. Sangwan and M.K.
Tiwari (eds), Land Use: Historical Perspectives. Focus on Indo-Gangetic Plains. Delhi:
Allied Publishers.
Vishnu-Mittre. 1972. Neolithic Plant Economy at Chirand, Bihar. Palaeobotanist, 21:
1822.
Walker, A. 1997. Indian Agriculture Asian Agri-History, 1(1): 121.
Wasson, R.J. 1984. The Sedimentological Basis of the Mohenjo-Daro Flood Hypothesis.
Man and Environment, 8: 8890.
Wasson, R.J. 2003. A Sediment Budget for the Ganga-Brahmaputra Catchment. Current
Science, 84(8): 10417.
Wasson, R.J., G.I. Smith and D.P. Agrawal. 1984. Late Quaternary Sediments, Minerals,
and Inferred Geochemical History of Didwana Lake, Thar Desert, India. Palaeogeography,
Palaeoclimatology, Palaeoecology, 46: 34572.
Wasson, R.J., M.S. Chauhan, C. Sharma, A.K. Singhvi, P. Srivastava and I.B. Singh (in
prep). Soil Erosion of a Durable Agroecosystem During the Last Two Millennia on
the Central Gangetic Plan.
Wasson, R.J., N. Juyal, M. Jaiswal, M. McCulloch, M.M. Sarin, V. Jain, P. Srivastava and
A.K. Singhvi (in prep). The Mountain-Lowland Debate: Deforestation and Sediment
Transport in the Upper Ganges Catchment.
Wessels, C. 1924. Early Jesuit Travellers in Central Asia 16031721. The Hague: Martinus
Nijhoff.
Wheeler, R.E.M. 1947. Harappa 1946: The Defences and Cemetery R37. Ancient India:
Bulletin of the Archaeological Survey of India, 3: 58130.
Whitecombe, E. 1972. Agrarian Conditions in Northern India. Vol. 1. The United Provinces
under British Rule, 18601900. Los Angeles: University of California Press.
Willcocks, W. 1930. Lectures on the Ancient Systems of Irrigation in Bengal and its Applications
to Modern Problems. University of Calcutta.
Wojtilla, G. (trans.) 1985. Kasyapiyakrsisukti: A Sanskrit Work on Agriculture. Acta Ori-
entalia Academiae Scientiarum Hung., 39: 85136.
51
Mesoamerica and the Caribbean Islands
A forty-year-old quote about the soils of the Maya realm remains accurate for
much of Mesoamerica and the Caribbean: A far more thorough examination of
the soil is needed before we can begin to interpret the ecological factors in Maya
culture history (Stevens 1964: 302). Indeed, we are still on the rising slope of an
exponential growth curve of knowledge of the history of soils (Dunning and Beach
2004b). In this region we have archaeology to thank for much of what we know
about the history of soils, because otherwise far less research would have been done
in nations that have little science infrastructure. Much of our knowledge of soil
history surrounds essential archaeological questions like subsistence and degradation
at ancient sites, and thus archaeology provides a foundation for understanding soils
at a time when they are changing faster than at any time in history. In this chapter
we show that soils share a long, complicated and dramatic history with humanity,
and that throughout history, especially over the last several decades, the highest
rates of change have often occurred with pioneers.
Rapid recent changes of soil landscapes have two principal drivers: popu-
lation change and land-use changes by groups that disregard ancient indigenous
soil management features. Mountjoy and Gliessman (1988) gave an example of
the misunderstanding of indigenous conservation technology in their description
of modern mechanised ploughing running over cajeta terrace systems, discussed
below. The ploughing destroyed what had been successful adaptation to farming
on steep slopes in Tlaxcala, Mexico by expunging canals that diverted water and
the cajeta basin that stored runoff, and this accelerated erosion on these steep slopes
(See Figure 4.1).
Soils and history in Mesoamerica and the Caribbean fall around two major
topics: human impacts over time and human soil practices or management over
time. At the outset, the first topic includes too little information about the current
52
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Chronological Overview
Recent soil history starts with the transition of the Pleistocene to the Holocene,
because of the climatic, geomorphic, hydrologic and vegetation changes that dra-
matically changed soils and the advent of humans in the Western Hemisphere
(though growing evidence supports a pre-Clovis and thus pre-Holocene presence
in the Americas at several more and less controversial sites). Soil studies in the
Teotihuacn Basin provide evidence for the changing climates of Late Pleistocene
53
Mesoamerica and the Caribbean Islands
and Holocene, in that caliches formed during dry periods in the Late Pleistocene
and argillic horizons formed during the warm and wet middle Holocene (McClung
de Tapia et al. 2003). In the Basin of Mexico, butchered Mammoth remains occur
with stone tools buried in peat soils that date to about 11,000bp (Morett et al.
1998), but the first direct use and therefore alteration of soil comes much later
with the advent of farming. In the Early Hunter Phase (11,000 to 7,000 years ago)
hunters and gatherers certainly used products of the soil, collected plants and seeds,
and managed these with their early domestication of squash and gourds and other
species (Piperno and Pearsall 1998: 312); but the first large impacts came later with
the Archaic Period (7,0004,000bp) associated with larger scale clearing by fire
and the growing bundle of diverse crops like corn, beans and squash (Piperno and
Pearsall 1998: 315). A remarkable group of cores and excavations near La Venta
in Veracruz shows evidence for the first maize pollen, the first sunflower seeds and
possibly the earliest Mesoamerican glyph (Pope et al. 2001; Pohl et al. 2002). Many
still think of Central Mexico as the home of maize, though Piperno and Pearsall
(1998, p.314), even before the announcement of 7,100-year-old corn pollen near
Veracruz, argued that the humid tropical lowlands were the major settings for
the origin and development of agricultural systems They see the development
of the maize and manioc here as the driver for the diffusion of swidden agriculture
from 7,000 to 5,000 years ago. Similarly, lake core studies in the heart of the Maya
Lowlands show maize pollen and disturbance indicators around 4500 bp or earlier
(Vaughan et al. 1985; Islebe et al. 1996; Wahl et al. in press). Such conclusions
hearken back to Carl Sauers more theoretical predictions of agricultural origins in
tropical wet and dry riverine environments (Piperno and Pearsall 1998, pp.1821;
Sauer 1952).
In the Caribbean Islands, (See Figure 4.1) the Archaic or Preceramic pe-
riod (Caribbean Island cultural Period 1), begins with the arrival of humans on
Cuba and Hispaniola between roughly 6000 and 5500bp (Rouse 1992). There is
mounting evidence that this first wave of colonisation originated on the Yucatn
Peninsula (Wilson et al. 1998). A second wave of Preceramic colonists spread
northward through the Lesser Antilles from northern South America, beginning
around 4000bp and reaching as far as Puerto Rico (Rouse 1992). Archaeological
assemblages from this period consist chiefly of lithics and often large quantities of
shells. Evidence for occupation on the islands typically occurs concentrated along
coastal margins near marine resources (Petersen 1997). Human disturbance is
also signalled by episodes of charcoal deposition in sediments in Lake Miragoane,
Haiti and Laguna Tortuguero, Puerto Rico between 5500 and 5000bp (Burney
and Burney 1994; Higuera-Gundy et al. 1999).
The next period in Mesoamerica is the Formative or Preclassic Period
(40001750 bp), during which the first environmental impacts on soils show up
from pioneer farming and land clearance by 3400bp as the start of the Maya Clays
54
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Antilles were populated by the Island Caribs most of whom practised a mixed
subsistence economy based on horticulture and marine resources.
After Conquest, populations declined precipitously throughout the Americas
but a new form of environmental change came from livestock ranching and burn-
ing to maintain grazing lands. Centuries, indeed millennia, of soil use by succes-
sive peoples make soils here a complicated palimpsest, which has led scholars to
divergent opinions about the causes of severe erosion in Central Mexico. In some
cases, soil erosion has been so severe that farmers have actively eroded what is left
of original soils, the cemented C-horizons that dominate surfaces, to tap softer
sediments beneath. But different studies focus their blame for upland erosion and
valley sedimentation on Pre-Columbian high populations using intensive agriculture
or on Post-Columbian depopulation and abandonment, the nefarious impacts of
livestock, and ham-fisted development schemes.
In recent times, rapid population growth and population displacements
have led to pioneering mentalities again and severe erosion in some parts of the
Maya regions and Central Mexico. Also, encouragement by state agencies and
international development to spur production has led to agricultural expansion,
especially since the 1950s. Much of the expansion was on to marginal lands, which
are too steep, too arid, or have other features that make them prone to degrada-
tion. This led to deforestation: about half Mexicos forests and much more than
half of its tropical forests have been removed (FAO 2005). Deforestation of course
greatly affected hydrology, soil erosion and soil nutrient depletion. For example,
gullying occurred on 65 to 85 per cent of dryland hillslopes in parts of Central
Mexico (Bocco and Garcia-Olivia 1992). Thus the grand soil problems that this
chapter will focus on are in many cases worse today than in most of the history of
soils of Mesoamerica.
A Parade of Paradigms
For Central Mexico, many historical sources from the Conquest period document
high populations and intensive cultivation in such forms as chinampas and terracing.
Moreover, the large abandoned sites of Central Mexico, like Teotihuacn, seemed
to show that ancient populations had also been high. Maya lowlands research
before the 1960s, however, believed that ancient Maya populations were low and
that soil manipulation was not intensive in nature. But broad, multidisciplinary
regional surveys since the 1960s have found more and more evidence for high
ancient populations and soil manipulations for intensive cultivation. This ushered
in a new paradigm of large population centres and the litany of intensive soil and
water management: dams, reservoirs, canals, terraces and wetland cultivation. There
have been important critiques of the methods for estimating high populations
(Bequelin and Michelet 1994, Rice and Culbert 1990), but settlement surveys
58
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
have announced high population estimates for more and more sites. Hence, the
current heterogeneous or mosaic model is that the Maya Classic had many large
and smaller centres that used many possible kinds of intensive and extensive forms
of soil management for subsistence (Dunning and Beach 2004b; Fedick 1996).
Now many studies by archaeologists, geographers, biologists, geologists and soil
scientists are testing the specifics of the Maya mosaic paradigm (Gomez-Pompa et
al. 2003). Therefore, we have organised this chapter along the lines of many forms
of evidence about the multiplicity of past soils and management.
Environments
Defining Middle America is an arbitrary process. We shall focus on the regions
from the basin of Mexico to Central America, spanning from about latitude 22
to 6 North and longitude 78 to 100 West. We also compare this region with the
islands of the Caribbean where the post-Columbian Old World/New World collision
first occurred (See Figure 4.1). The environmental diversity of Middle America is
impressive in geology, climatology, ecology, soils and human environmental history.
Geologically, the region ranges from the Mexican Plateau, to the high and active
volcanic peaks at the spine of most of this landmass, to metamorphic mountain
ranges, to deeply incised canyons, to a fringing coastal zone, to a vast, flat carbon-
ate plain, which is intruded by a small granite batholith in the Maya Mountains
in Belize. Geological diversity is only matched by climatological and ecological
diversity. Middle American ecosystems and climates range from near deserts to
tropical forests and montane coniferous forests. The region lies mainly in the Trade
Winds, with occasional intrusions of Westerlies mid-latitude cyclones and seasonal
dominance of the subsiding air and dry conditions of the Subtropical High or the
precipitation of the Intertropical Convergence Zone. These large systems and com-
plex topography lead to tropical wet and dry climates with high regional variations
in precipitation, humidity and temperature, and the periodic occurrence of severe
droughts (Hodell et al. 1995; 2000; 2001; Haug et al. 2003).
limestone terrain. This landscape arose differentially from the ocean since the late
Cretaceous (Donnelly et al. 1990), which caused a chronosequence of soils to form
based on how long they have been above sea level. One chronosequence lies in
the northwest Yucatn, where concentrically older and older soils occur outward,
and inland, from the Chicxulub impact crater, since the centre has been the most
recent part to emerge from the sea (Pope et al.1996).
Five common soils that form on the limestones across Yucatn are entisols
and inceptisols, thin and young soils that form in recently deposited material;
rendzinas or rendolls (or pusluum or boxluum in Yucatec Mayan); alfisols or terra
rossas (or Kankab in Mayan terms); vertisols; and histosols. Histosols or peats and
mucks occur all over the world in low-lying areas with water tables near the surface
(see Chapter 5). The inceptisols, entisols and rendolls are young, fertile, thin, clayey
soils that range up to about 50cm in depth and are differentiated by thickness
and horizon formation. In the Maya Lowlands, the alfisols (Ustalfs or Kankabs)
are older, red, iron-rich, less fertile, generally thin, clayey soils. The rendolls form
from organic matter, high amounts of carbonate and residual silicate minerals
from the limestone parent material, and some wind and volcanic deposition. The
alfisols form from the same materials but the organic matter generally decomposes
more, little carbonate is left, and more silicate clay and iron minerals build up over
greater periods of time (Beach 1998a). Vertisols are soils that form from expand-
ing and contracting clays, which also form surface cracks and ridges called gilgai.
Vertisols form in deeper profiles in karstic upland sinks or poljes such as the bajos
discussed in Chapter 5 (Beach et al. 2003a). In all of these soils, silicate minerals,
which dominate in many soils elsewhere, occur as impurities within limestones
that are dominantly calcite (CaCO3) (Isphording 1984; Beach 1998a). The CaCO3
dissolves, leaving little mineral matter for soil formation. Thus these Yucatn soils
are usually thin, relatively high in organic matter, and contain the residual silicate
minerals, especially kaolinite, from the limestone, winds or volcanoes. Higher than
expected organic matter in these tropical soils may be caused by fine-sized secondary
carbonates impeding decomposition (Shang and Tiessen 2003).
Williams and Ortiz Solorio (1981) note that the view is widespread in the
Mexican countryside that experts are outsiders who know more than peasants about
everything except the soil, about which they are the experts. Pulido and Bocco
(2003) show this expertise in a rural indigenous community in Central Mexico,
in that the community values its soils and their management, which translates into
sustainable soil management with relatively low gullying on very erosion-prone
slopes. Thus ethnographic studies provide invaluable knowledge and contemporary
insights into history, and we are fortunate to have ethnographies, histories and two
extant codices for the Aztecs. These codices contain glyphs that provide soil data
on 1100 agricultural fields in the Basin of Mexico. Williams and Harvey (1997,
p.30) stated that these complex, written Nahuatl soil classifications are unique in
60
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
the Americas and they knew of only one other parallel in fifth-century bc China
(see Gong et al. 2003). The classification in the Codice de Santa Maria de Asuncion
shows 104 variants for soils in 200 hectares, to which peasant farmers today apply
four classes, and modern taxonomy ascribes five soil phases. The 104 variants prob-
ably represent 18 taxa and three separate class levels. These descriptions include
a number of aspects that relate to soil productivity such as texture (e.g., sandy or
clayey); landscape positions (e.g., alluvium or hillslopes); irrigation; and colours
(e.g., black or yellow). The most productive of the generic taxa was Atoctli, which
refers to alluvium of several kinds. Another interesting taxon is tepetlatlalli, which
refers to the well-known caliche soil C-horizons, known as tepetate, that become
exposed after about 70cm of erosion. This Codex indicates a large area of tepetate,
and thus soil erosion, before the mid-sixteenth century ad (Williams and Harvey
1997, p.34). Moreover, Sahuguns Florentine Codex from the same period even
discusses the practice of purposely breaking up this eroded subsoil to make tepetlalli
soils for planting.
What we know of the soil history of the Maya realm is less than what we
know for Central Mexico, and it is based on ethnography and archaeology rather
than texts. In most of the Petn region only light reoccupations by Maya groups
disturbed the many centuries of abandonment that followed the Terminal Classic
period (Palka 1999). These groups took with them their evolved soil knowledge
(see Nations 1979). In the Maya-language areas of the Chiapan and Guatemalan
Highlands complex indigenous soil knowledge persists (See Figure 4.2). One of
the most distinctive features of these knowledge systems is their categorisation of
various soils as either hot or cold (Maffi 1999). While these terms refer in part to
the amount of insolation experienced by various soils, the terms also have significant
overtones with regard to fertility and undertones of a political nature due to the
historic displacement of many Maya groups on to poorer cold land.
Many of todays Maya farmers of the Lowlands are shifted cultivators from
other areas, often the Highlands, and they have either recently adapted their soil
classifications or brought them from other places with very different soil types.
Indeed, one study of the Kekchi Maya of the Petn, who had recently immigrated
from the Highland Alta Verapaz, showed that they classified about 60 per cent of
their new land as trash land (Dunning 1992; Dunning and Beach 2004a), but
they began to recognise greater diversity over time. The Kekchi of the Petn have
four basic levels of soil that differentiated warm and cold, texture and drainage,
and color and depth (Dunning and Beach 2004a). Earlier Kekchi emigrants from
Alta Verapaz to another tropical lowland near Lago Izabal, Guatemala also replaced
their cold earth taxa with a drainage-based class (Carter 1969).
Yucatec farmers have been in place for much longer, and have direct con-
nections to the historical and prehistorical Maya of their region. They also have a
significant taxonomy which describes an impressive range of prospective harvest
61
Mesoamerica and the Caribbean Islands
zones. Their classification defines seven main farmable landscape taxa and numerous
amending taxa. The taxa also refer to drainage, colour, topography and texture; yet
they are also replete with generations of accrued knowledge about what crop types
grow best in particular soils (Dunning and Beach 2004a; Dunning 1992).
Maya farmers have also influenced soil formation over three millennia.
Their presence and activities have consciously and unconsciously created middens,
fertilised areas, depleted areas, and formed new soils. Their largest impacts have
been on erosion of uplands and on the large areas of soils created by hundreds
of abandoned Maya cities. Some literature showed that Maya sites had high soil
quality and suggested this had been a pull factor for settlement. This may be cor-
rect, but we can say empirically that Maya activities at some sites have improved
soils. For example, fertile mollisols around the colossal site of Chunchucmil in
the northwest Yucatn are largely the creation of ancient limestone, plaster, floors
and middens. Soils that formed with none of these are alfisols which are more
leached, thinner and less fertile, and contemporary Yucatec farmers often ignored
them, planting directly on and around ancient architecture in the anthropogenic
mollisols (Beach 1998a; Dahlin et al. in press). This pattern occurs on many of
the so called Garden Cities of Yucatn, where the often black anthrosols lie next
to and on top of older red soils. These Maya soils have many of the attributes of
the Amazons terra preta: dark colour, higher fertility, charcoal, many artefacts and
active use, including mining, by farmers (Woods and McCann 1999; Lehmann et
al. 2003; see also Graham in press).
Since the 1990s, more scholars have used a full range of analytical tech-
niques such as soil micromorphology, carbon and oxygen isotopic studies, thin
layer chromatography, gas chromatography/mass spectrometry, inductively coupled
plasma-atomic emission spectrometry (ICP/AES),and inductively coupled plasma
mass spectrometry (ICP/MS) with a variety of extractants. Several studies in Mesoa-
merica and elsewhere have found that archaeological features have distinct chemical
signatures from natural soils and that floors, hearths, gardens and pigments can be
distinguished based on this multi-elemental technique (Middleton and Price 1996;
Wells et al. 2000; Parnell et al. 2002; Cook et al. 2006).
Biochemical techniques are also expanding, but again few studies have been
published thus far, though biomarkers hold large potential for answering questions
about ancient fertilisers and human activities on soils (Bull et al.1999). For example,
current projects are attempting to work out ancient marketplaces and fertilisers
to get at old questions about how ancient cities functioned and how enough food
could have been produced in poor soils (Fernandez et al. 2002; Fedick 2003; Dahlin
et al. in press). Barba and colleagues working in Mexico have analysed a variety of
elements and biomarkers (e.g., albumin, carbohydrates, fatty acids) in short distance
variation in archaeological and contemporary sites (Barba et al. 1987; Manzanilla
and Barba 1990; Barba and Ortiz 1992).
Carbon isotope ratios through a vertical sequence of depositional soil profiles
provide another approach to the question of what grew on paleosols because they
reflect the amounts of C3 and C4 plants that formed different types of soil organic
matter (Fernandez et al. 2005). Most of the tropical forest broadleaf plants have
13C ratios of -26 to -30 per mil, whereas the C4 plants such as many grass species
have 13C ratios of -12. Many of the surface forest soils have 13C ratios similar
to forest species, but buried soils, some dateable to the ancient Maya period have
ratios that are much closer to C4 plants. Interestingly, one study even found that
a contemporary milpa (swidden), which had grown maize for 20 years still had a
ratio similar to the forest soil, which probably indicates that soils with 13C ratios
similar to C4 plants must have had maize and other C4 plants growing for long
periods of time (Webb et al. 2004; Fernandez et al. 2005).
Fossils supply another traditional proxy for past soil uses and alterations.
Here many studies have used the full range of fossils, including pollen, phytoliths,
mollusk shells, diatoms and other skeletal remains (McClung de Tapia et al. 2003;
Dunning et al. 2002; Hansen et al. 2002; Fedick 2003). As with biomarkers and
chemical signatures, these preserve differently in different environments and they
may not be very representative of past soil environments. As with all studies to re-
construct the past, multi-proxy approaches are the best, though the most expensive,
because they provide multiple lines of evidence that may converge or diverge.
64
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Soil Catenas
A common method for understanding soils is to study soils where all variables but
one are constant, and to study how they differ by this one factor: climo-, bio-, topo-,
litho- and chronosequences, discussed above (Birkeland 1999). A catena (similar
to a toposequence) is a series of soils down a slope that vary due to drainage and
topographic position (Jenny 1980; Birkeland 1998; Sommer and Schlicting 1997).
Only a handful of projects have studied toposequences or other sequences across
this region and they become important to history when they provide evidence for
human induced change.
A comparison of enough soil profiles at each slope segment with natural
versus altered land uses can provide a general picture of soil losses. For example,
Furley (1987) compared the changes that occurred because of one milpa cycle be-
tween 1970 and 1981 on a catena near Belmopan, Belize. He concluded that soil
losses were so high from one milpa cycle that soils could be eroded to bedrock in
as few as 4 milpa cycles. Other studies of catenas in the Maya region using several
methods echoed Furleys. Soil losses were very high in every method of evaluation
and soils were truncated by nearly one-half on average over a decade and in some
cases, erosion removed entire soil profiles in the same time period (Carlos Donado
1996; Coultas et al. 1997; Beach 1998b; Fernndez et al. 2005; Beach et al. 2006).
Some studies have also attempted to compare catena soil losses to estimates from
the Revised Universal Soil Loss Equation (RUSLE) on these tropical, Mesoamerican
slopes (Beach 1998b; Millward and Mersey 1999).
of organic matter and sparse residual silicate matter from their parent materials.
Because the soils are rocky, patchy and thin, modern, mechanised agriculture often
fails, though traditional Maya digging sticks eke out yields.
A parallel contention of human-induced change by the ancient Maya is that
the savannas and lateritic (oxisol) soils of the central Petn (around La Libertad,
Poptun and elsewhere) were induced by repeated burning of forests and the soils
subsequently laterised by intensified leaching and weathering (Stevens 1964, p.299).
Unfortunately, little modern research has taken up this interesting question (see
Carlos Donado 1996).
Two giants of early twentieth-century Maya archaeology, Morley (1946) and
Thompson (1954), took up the earlier suggestion of agricultural exhaustion as an
explanation for the Terminal Classic collapse in the Petn and the subsequent rise
of the Post-Classic centres of the Yucatn (Dunning and Beach 2004b). Stevens
(1964, p.302) also perpetuated the notion that soil depletion caused the Late Clas-
sic Maya collapse in an era when several studies linked soil depletion with collapses
of ancient societies (Bennett 1945; Lowdermilk 1953; Morley 1956, p.71). These
ideas continue to the present in the assertion by Healy et al. (1983) that terrace soils
degraded to toxic effects at Caracol and soil erosion and depletion contributed to
the collapse of Copn (Abrams and Rue 1988; Wingard 1996), which Diamond
(2005) takes up in more popular literature. The collapse scenario for Copn was
based on pollen cores that showed high soil erosion rates and forest clearance in
the Late Classic (Abrams and Rue 1988). Deforestation and accelerated erosion
spread up into foothills and upper slopes as the population expanded, cut more
wood, intensified swidden and decreased fallow times. Wingard (1996) built on
the soil erosion/collapse scenario for Copn by using the Erosion Productivity-
Impact Calculator (EPIC) computer model to simulate very high soil erosion rates
and fertility declines with population expansion. The EPIC model estimated that
fertility declines alone from the Late Classic would have required one hundred
years for recovery.
Central Mexico
Many similar soil narratives exist for Central Mexico as for the Maya Lowlands,
except that most soil practices and changes started at least as early and continued
later in a more diverse environment. We have much more information from surviv-
ing indigenous soil use, taxonomy, agriculture and conservation in Central Mexico
than in the Maya world since we have extant indigenous documents as well as
many colonial documents.
The Central Highlands is a diverse landscape of cultures and ecosystems. The
region includes central basins, mountain slopes, piedmonts and coastal plains. The
Mesa Central consists of high, semi-arid, frost prone basins like Puebla, Touluca,
Oaxaca and Mexico. Here the populations were dense, the cultures diverse and
66
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
history long, with maize agriculture, pottery and sedentism stretching back farther
than 4,500 years. These create a very complicated landscape for uncovering soil
history, but the region is rich with studies that break down along the usual lines of
soil history: soil erosion, folk taxonomies and soil management.
Studies of soils and soil erosion here range from process-based work on gully
formation (Bocco 1993) to syntheses that attempt to make some broader sense
of these complex interactions of soils and humans. For example, the special quin-
centennial issue of the Annals of the Association of American Geographers, edited by
Karl Butzer (1992a and b), had several articles that reviewed the massive alterations
present in these landscapes. These are often divided into studies that highlighted
degradation induced by Pre-Columbian or European settlement. Denevan (1992)
focused on severe Pre-Columbian Aztec erosion in Cook (1949, p.86) and the
tepetate, the Pre-Columbian exposed petrocalcic C-horizon that was such a wide-
spread surface (Williams 1972; McAuliffe et al. 2001). Likewise, Butzer and Butzer
(1993) working in the Central Mexican Bajio found no evidence for early livestock
impacts being greater than those of Pre-Columbian times. These and many other
works thus formed the basis for the Pristine Myth synthesis in the early 1990s,
which showed that Pre-Columbian societies had severe environmental impacts and
intensive land use. Nonetheless, many studies focus on European settlement as the
degradation driver, especially in the last fifty years.
Another useful synthesis by C. Rincon Mauntner (1999, pp.5913) de-
veloped nine erosion themes: (1) humans caused vast soil erosion (Cook 1949,
pp.224; Heine 2003) and population density correlated with soil erosion (Cook
1949, pp.1486; Heine 2003); (2) climate and hydrologic change caused erosion;
(3) soil erosion was too widespread to be caused by humans; (4) indigenous farmers
are/were conservationists and most erosion is due to European land use changes;
(5) soil erosion was linked to very recent socioeconomic processes; (6) severe soil
erosion was linked to particular lithology (Kirkby 1972; Stevens 1964); (7) soil
erosion was deliberate and induced since the La Natividad Phase (ad 10001530)
(Sahagn 1577[1900]; Spores 1969, pp.5634); (8) importance of European live-
stock causing erosion (Melville 1994); and finally that (9) depopulation exacerbates
erosion (Spores 1969; Kirkby 1972). Many of these themes run through soil erosion
research around the world, and recall the complicated and heavily studied question
of arroyo formation in the Southwestern United States (Bryan 1925; Tuan 1966;
Cooke and Reeves 1976).
Often several of these nine themes run through an individual study and
probably influenced later Maya studies. For example, Spores (1969) presented a
soil erosion model that is similar to the model at Copn: low soil erosion in the
Formative and Classic Periods because agriculture focused on valleys, now buried
in many cases by 1.52 m of alluvium. Farmers started to move up slope during
the Las Flores Phase (ad1501000) and after, and despite building cross-channel
67
Mesoamerica and the Caribbean Islands
Overview of Erosion
These basic facts underscore three principles of land degradation through time
in Mesoamerica: there was a diverse range of impacts; pioneer farming and land
clearance produced more severe degradation than earlier work reflects; and con-
68
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
servation developed with growing societies and florescence in many, but not all,
Classic sites. This is borne out in many of the Maya Lowland examples, where soil
erosion starts in the Early Preclassic, conservation develops to its fullest extent in
the Classic period, and massive alteration and soil erosion returns in the last few
decades. This is also borne out in the Highland Mexico at Lake Patzcuaro, where
the authors argued that most degradation occurred with site construction and
again with abandonment in the Colonial period (Fisher et al. 2002). The story
is somewhat different around the Basin of Mexico, where Cordova and Parsons
(1997) argue that soil erosion accelerated with depopulation and abandonment of
the rural landscape in both pre- and post- Aztec (Colonial) times. Thus half of the
model also works for much of Mesoamerica in that erosion declined with active
soil conservation during periods of high rural populations, but there seems to be
a major difference between the moist, forested Maya Lowlands and the semi-arid
steep lands of Central Mexico. Soil erosion essentially stops with abandonment and
rapid reforestation in the Maya examples, whereas the highest erosion sometimes
occurs with abandonment in Central Mexico. Significant alteration still comes with
pioneer settlements in most studies of both regions. The debate, however, still rages
about early post-settlement degradation from livestock, though recent rates of ero-
sion from more intensive land uses on steep lands are the most severe yet.
Lake Patzcuaro
One of the most interesting fronts of the Pristine Myth debate that still goes on
today is research by different teams at Lake Patzcuaro. Indeed, one of the death
knells of this myth was a study of cores from this lake in Michoacans Central
Mexican Altiplano that found that erosion rates were at least as high during Late
Preclassic and the Post-Classic as they were after the Spanish conquest (OHara et
al. 1993; Butzer 1993). These findings made the authors question whether a return
to indigenous methods as an approach to development was appropriate, given their
findings that indigenous methods caused at least as much erosion as plough- and
livestock-induced erosion. Fisher et al. (2003) in a study of the same lake using
cores, trenches and exposures came to different conclusions. They argue that the
Preclassic degradation was caused by initial clearance rather than agriculture, that
high population is inversely related to soil erosion, and that the post-European soil
70
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
erosion was tied to a loss in labour from the many contagions that accompanied
the Spanish conquest.
In summary, since the late 1980s, many commentators have framed the
discussion of human-induced soil change into debates over the Pristine Myth,
which has pointed out once again how theory moulds perception of environmental
change. What all of these erosion studies in Central Mexico underscore is our need
for more refined soil histories using multiple lines of evidence and dating. Thus
far, we have different chronologies of human impacts on soils, but one common
sequence occurs as follows: (1) Holocene equilibrium soil development into the
Early Preclassic; (2) soil erosion starting in the Preclassic, based on dates of the sur-
faces of buried soils and sediment cores; (3) soil development with less disturbance
from sometime later in the Preclassic to the start of the Late Classic; (4) increased
soil erosion in the Late Classic, but mitigated by widespread soil conservation; and
(5) recovery and topsoil formation in the Post-Classic. The extensive terracing that
diffused widely in the Late Classic did not expunge soil erosion but it may also have
forestalled any widespread Malthusian menace that some scholars suggest for some
sites. Yet, some sites such as Cancun and Copn did have intensive soil erosion
and sedimentation and little soil conservation.
Terracing
Agricultural terracing stood out to Maya archaeologists as early as the 1920s, but
they did not recognise its implications for labour or as a form of intensive agri-
culture (Beach et al. 2002). Many studies have found that terracing across ancient
Mesoamerica was a common and successful adaptation to the regions steep, tropi-
cal environments and eroded or skeletal soils. Donkin (1979, p.131) writes that
terracing occurs mainly in arid and semi-arid regions, though we know now that
it also occurs in extremely wet areas. Indeed, in some areas with about 2,500mm
of precipitation, we have found terraces that continue to operate after 1200 years
under the active geomorphic surfaces of tropical forests and milpas (Bocco 1991;
Beach and Dunning 1995; Beach et al. 2002). In the early 1990s, there were still
few real studies of terraces based on excavation and survey (e.g., Turner 1974 and
Healy et al. 1983), but since then many studies have remedied this (e.g., Doolittle
1990; Dunning and Beach 1994; Fedick 1994; Treacy and Denevan 1994; Beach
and Dunning 1995; Beach 1998a; Dunning et al. 1997; Beach et al. 2002). Exca-
vation and survey of terrace systems are keys to understanding their construction,
function, and soil evidence for past crops and cropping techniques.
The remains of ancient terraces include stone walls and support, soil sequences
and artefacts; all remnants can provide dates for construction. Donkin (1979,
p.18), in his book on terracing in the Americas, wrote that terracing started by
about 2500bp in the Andes and Central Mexico. Since accelerated erosion started
before 3000bp, the need for terracing started earlier. Indeed, since Donkins work,
71
Mesoamerica and the Caribbean Islands
research in Northern Mexico, the Maya Lowlands and Central Mexico all points to
terrace origins around 1000bc in Northern Mexico. Hard and Roney (1998) and
Hard et al. (1999) provide strong evidence for a sizable Archaic terrace group that
dates to about 1050bc, based on radiocarbon dates for maize and squash remnants
and Late Archaic artefacts. In the Maya Lowlands, Hansen et al. (2002) reported
late Middle Preclassic (c.600400bc) terracing at Nakbe, Guatemala, and in the
Central Highlands, Mountjoy and Gliessman (1988) stated that cajeta terracing
starts by about 1000bc. Despite the evidence for 3,000-year-old terracing, most
terracing started in the Maya World only in the Classic Period (Beach et al. 2003a),
whereas terracing is widespread geographically, culturally and chronologically across
Central Mexico (Donkin 1979).
Worldwide, terraces function to provide a planting surface, maintain
soil moisture favourable to crop growth by managing water infiltration, reduce
soil erosion through slowing overland flow, catch or mound soils on eroded or
thin-mantled slopes, and redirect water toward or away from lower lying fields or
water sources (Doolittle 1985 and 1995; Beach et al. 2003). Many studies have
interpreted evidence from excavation for each of these functions, and many others
have developed taxonomies of terracing (Treacy and Denevan 1994; Beach and
Dunning 1995; Beach et al. 2002). Such factors as landscape position, size, con-
nection to drainage and irrigation, and materials form the bases for classifications.
Poorly built or unmaintained terraces, however, are prone to mass wasting and
gully erosion, because gullying by piping or waterfall action can undermine them
(Beach and Dunning 1995).
What we know of construction materials reflects the limitations of methods:
excavation and mapping on the active tropical slopes of Mesoamerica show well-
made dams of heavy stones, but these may be the more persistent remnant anchors
of a landscape that also included soil and vegetative berms. We know, for example,
from the Basin of Mexico, vegetative walls such as maguey hedges (Sp. bancales and
in Nahuatl metepantlis) served as semi-terraces that collected sediment around them
over time (Palerm and Wolf 1957; Donkin 1979; Evans 1992). In these Mexica
lands, aqueducts and canals irrigate the soil beds of rock-walled contour terraces
(Cordova and Parsons 1997).
Interestingly, some terraces have unknown and perhaps unknowable func-
tions in soils because they do not follow any usage rules we can ascertain. In north-
western Belize, for example, chert berms have erratic sizes and as often run parallel
or diagonal to slopes as normal to them. The only clues are their Late Classic dates
and largely chert construction. Perhaps these were field walls or rock piles removed
for tilth and reapplied for mulch (Beach et al. 2002).
In the Maya Lowlands the greatest extent of terracing occurs around the
Rio Bec in Mexicos south-central Yucatn Peninsula and around Caracol on the
flanks of the Maya Mountains in Belize. Large scale terracing also occurs in Ver-
72
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
acruz, Mexico. Sluyter and Siemens (1992) reported that more than 1000km2 of
terracing are adjacent to the well-known wetland fields in Veracruz (see Chapter5).
These lie outside the Maya Lowlands, but the authors date them to a period that
corresponds broadly to the Maya Classic. Turner (1974) estimated more than
10,000km2 of terraced lands in the Rio Bec region, and Healy et al. (1983) and
Chase and Chase (1998) described Caracols incredibly widespread terrace systems,
which they argue must have been centrally controlled. Williams (1990) posits that
many terraces evolve with little planning as only localised responses to landscapes,
and many are built free-standing to dam eroding sediment and create a planting
bed (Williams and Walter 1988). Beach et al. (2002) have also argued for more
localised control near La Milpa and Dos Hombres in Northern Belize, though
some terracing suggests broader watershed planning because the terraces connect
to diversion channels and possibly intensively farmed bajos. Work in Tlaxcala in
Central Mexico shows more integrated agroecosystems with sloping terraces that
divert runoff into cajetas or basins, which farmers clean out and reapply to fields
(Mountjoy and Gliessman 1988). Others studying the Aztec suggest that Late
Post-Classic agricultural terracing was the result of household level decisions under
growing populations (Smith and Price 1994). There is danger in reading central
planning of soils and water management into an archaeological landscape because
piecemeal building over time may appear the same as central planning, especially
after many centuries (see Doolittle 2000, p..300301).
Whether or not Mesoamericans centrally planned these soil conservation
features, we are learning more and more about soil-terrace systems. Terracing
occurred widely in Mesoamericas lowlands and highlands, but did not occur in
some sloping areas that warranted it. For example, many parts of Maya Mexico
and Guatemala and seemingly everywhere in Belize had Classic Period terracing
Caracol, Xunantunich, La Milpa, Dos Hombres and the Petexbatn but there
was surprisingly little terracing at Copn, Honduras and around Tikal, Guatemala
(Beach et al. 2002). Studies also suggest terracing was often a successful and sustain-
able agroecological system, though work at the same group of terraces at Caracol
drew opposite conclusions on this point (Healy et al. 1983; Coultas et al. 1993).
Research is also beginning to recognise the crops of ancient terracing, borne out
in the latest chapter of Caracol terraces: carbon isotopes that suggest maize and
other C4 grass species (Webb et al. 2004). Not surprisingly, studies have drawn
more Malthusian conclusions about the sites without terracing such as Copn,
with population-driven soil erosion and depletion (Wingard 1996), whereas high
populations elsewhere such as in Central Mexico correlate with terrace construction
to restore soils that had been truncated to their tepetate parent materials (Williams
1972: 626).
73
Mesoamerica and the Caribbean Islands
Field Walls
Albarradas or field walls occur across many sites in the Yucatn, especially Chun-
chucmil, Coba, Dzibilchaltun, Yalahau and Mayapan. Since these are such promi-
nent parts of the sites, and even occur near the site centres, they must have some
importance. In many cases they enclose a small area of a mound group and adjacent
areas with no structures and often mediocre soils. The simple explanation is they
were property boundaries of one kind or another, but chemical studies have not
identified a proxy of soil use as yet.
At Yalahau, in the northeast Yucatn, however, Fedick and others have ar-
gued that low lying walled areas may have been for growing periphyton, microbial
wetland mat communities dominated by cyanobacteria, as a superfertiliser that also
has pesticide properties (Fedick et al. 2000). These research teams report evidence
of possible periphyton use in the form of wetland sponge spicules and mollusc
species in the forest soils near sites, and they are testing other evidence (Palacios-
Mayorga et al. 2003). This should be a prime use of biomarkers as proxy evidence
for wetland microbial inputs into upland soils.
parts of Mexico, the ridges, called camellones, still echo the Nahautl word cuemitl.
In contrast, the thin rocky soils of the northwest Yucatn cannot be manipulated
thus, but farmers still focus on cropping the higher, naturally mounded karst uitzes
(Beach 1998b).
Kitchen Gardens
Estimates vary for the importance of kitchen gardens or solares to Mesoamerican
subsistence and their importance on soils. Kitchen gardens are a form of inten-
sive or special soil management. As such, they usually require deeper soils and
anthropogenic inputs, and may develop anthrosols that we can detect in ancient
landscapes. These are probably partly the cause of the anthropic mollisols at sites
like Chunchucmil and Mayapan (Beach 1998a).
There have been many studies of solares in Mesoamerica, and some show
intensive production with high inputs and polycultural techniques that attempt
to maximise land area by vertically arranging crops into multiple canopies (Atran
2003; Atran et al. 1996; Barrera et al. 1977; Killion 1992; Nations and Nigh
75
Mesoamerica and the Caribbean Islands
1980). For Central Mexico, Williams and Harvey (1997) suggests that kitchen
gardens ranged from 0.3 to 0.9ha and thus contributed greatly to subsistence. Yet
for the Maya Lowlands, ethnohistorical analogy suggests more humble production.
Caballero (1992) wrote that polycultural gardens produce mainly nutritional and
dietary supplements, medicinal herbs and plants used as ornamentals and ritual
objects, rather than bulk staples, and he concludes that kitchen-gardens produce
only about 11 per cent of the average domestic calorie intake. Archaeological evi-
dence, however, indicates that prehispanic gardens may have contributed more to
subsistence, though as landscapes became more and more urbanised the distinction
between former infields and kitchen gardens likely became blurred (Folan et al.
1979; Smyth et al. 1995).
river came near there, and in such a place they made a fire, and afterwards the river
moved farther away. Soil accumulated there, as the rains brought it down from the
hills and covered the site. And because this could not happen except by the passage
of many years and most ancient time, there is therefore a strong argument that the
people of these islands and the mainland are very ancient (Las Casas 1990).
There has also been limited scientific study of ancient erosion and deposition, in
part because of the existence of only a small handful of freshwater lakes suitable
for paleoenvironmental investigation. Several cores have been analysed from Lake
Miragoane, a cryptodepression in Haiti (Brenner and Binford 1988; Higuera-Gundy
et al. 1999). These cores show evidence of deforestation and moderately increased
sedimentation after about 1200bp. However, the record of pre-Hispanic deforesta-
tion and sedimentation is comparatively minor in comparison to that in colonial
times and especially the horrendous problems of the past 50 years.
By 1300bp, farmers on Puerto Rico were sufficiently aware of soil erosion
that they started building agricultural terraces. Although pre-Hispanic agricultural
terracing has been reported in Puerto Rico since the 1930s, systematic investigation
is more recent (Ortiz Aguil et al. 1992; Oliver et al. 1999). The onset of terracing
appears to be linked to the expansion of cultivation away from valley floors and coastal
plains onto the lower slopes of mogotes in the island interior. The large majority
of terraces investigated to date are stone-faced, dry slope, contouring terraces, but
there are few examples of cross channel terraces and diversion weirs.
In drier areas of eastern Hispaniola ditch irrigation was employed in pre-
Hispanic times, but the extent and dating of this practice remains poorly under-
stood (Tabio 1989; Veloz Maggiolo 1976). By far the most widely practiced form
of pre-Hispanic intensive agriculture was the construction of thousands of small
raised beds or conucos (Sauer 1966). Typically conucos were constructed as a series
of small circular earthen mounds on the order of 25cm high and one metre in
diameter, sometimes associated with shallow irrigation ditches. Often green mulch-
ing was used to help conserve the soils and moisture in the mounds. While cassava
was the primary crop grown in conucos, they were typically used for mixed-crop
cultivation also including sweet potato, arrowroot, maize, bean, squash, peppers,
peanuts, cotton and tobacco.
Conclusions
In our review of the soils history of Mesoamerica and the Caribbean Islands we have
attempted to find patterns and trends for vastly different environments. One clear
theme from many places is the advent of human impacts in the form of Maya clays
or eroded sediments over paleosols or organic lacustrine sediments. In the Maya
Lowlands, the possible advent of human impacts comes first with change of more
forest to more savanna species in the Petn of Guatemala about 5610bp. This may
77
Mesoamerica and the Caribbean Islands
just as well be caused by climate drying as human alteration, but by 3400bp increased
sedimentation of inorganic Maya clays and pollen of economic and disturbance taxa
into depressions provides convincing evidence that deforestation and soil erosion
had started (Brenner et al. 2003). Soil conservation and soil manipulation in the
form of terracing also started in the Preclassic in the Petn and Belize (Hansen et
al. 2002; Beach et al. 2002), but most conservation was slow in developing and
did not become widespread till the Classic and, especially, the Late Classic. Thus
the impacts of pioneer agriculture in the Preclassic aggraded many depressions
and valleys (Beach et al. 2003a), but Late Classic soil conservation started a trend
toward sustainable agriculture. Nonetheless, several important Late Classic Maya
sites have little evidence of conservation and much evidence of erosion (Beach et
al. 2003b). The loss of soil, however, stopped dramatically over much of the Maya
Lowlands with collapses in the ninth and sixteenth centuries: forest returned, soil
formed, and sedimentation slowed down to an organic trickle.
In Central Mexico, human induced soil alteration also started in the Pre-
classic, in the form of site construction and clearing for agriculture. Likewise, soil
conservation started early, both in Central Mexico and Northern Mexico, by the
Early Preclassic. Maize domestication must have started somewhere in this zone
before 7000 bp (Pope et al. 2001), but evidence for major human-induced land
use change that might have accompanied maize farming comes later, possibly as
farmers moved up the slope. No single collapse quiets the environmental record, as
occurred in the Maya Lowlands, and indeed the depopulation of before and after
the Aztecs brought arguably more environmental impact because of the collapse
of native populations that had maintained conservation features and the spread of
livestock-induced erosion. The legacy of erosion on the upper piedmonts and hill
slopes of Central Mexico are widespread, truncated soil profiles called tepetate, and
they still erode at nearly five times the rate of the areas with relict soils. In both
Central Mexico and the Maya Lowlands, new pioneers have caused modern soil
erosion from mechanised, modern techniques and ignorance about or indifference
toward their newly deforested but ancient Mesoamerican fields. Some studies, though,
have shown extremely unsustainable levels of erosion, such as the plot studies along
the Pacific Coast of Mexico that show maize and grass soil loss as 70 and 49 t/ha/yr
(Maas et al. 1988), soil erosion and GIS modeling studies to optimise the timing of
land use actions (Millward and Mersey 1999), and a 137Cs study under undisturbed
deciduous forest that showed a soil loss of 13.2 t/ ha/yr (Garcia-Oliva et al. 1995).
Soil erosion is particularly a problem in tropical soils because top soil layers hold a
high percent of the major soil nutrients (Lal 1990). There can be little doubt that
Mesoamerica has severe rates of soil erosion today, but we lack enough of these
well designed and instrumented studies to give a full assessment.
In the Caribbean Islands, particularly the Greater Antilles, agriculture was
slower to develop and most sedentary populations concentrated along coastal
78
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
margins with relatively low erosion risks. However, as populations moved onto
sloping inland terrain, the need to conserve soils became obvious and a tradition
of terracing evolved. Unfortunately, this conservation knowledge was lost when
indigenous populations were decimated in early Colonial times. The modern record
of catastrophic soil loss in such places as Haiti stands in sharp contrast.
The rest of what we know about the history of soils in Mesoamerica has to
do with soil manipulation to grow crops, ethnopedology studies, chemical assays,
and archaeological excavation to discover what ancient people might have been
doing on their lands to feed their hordes. We know that ancient urban and rural
populations were extremely high, especially in Central Mexico (Williams 1989)
and the ancient Maya Lowlands (Turner 1976), where agricultural productivity
today is low and soils depleted. This riddle motivates our soil science today to try
to answer questions of ancient subsistence, which might even provide us with some
insights toward improving contemporary farming through agroecology (Gliessman
1998). We have started to gain insight into ancient soil knowledge, first from eth-
nographies of farmers about soils, combined in Central Mexico with codices that
record remarkable Aztec soil knowledge. We have only the ethnopedologies from
indigenous people and no such written records from the rest of Mesoamerica, and
we can only ask in vain whether this Aztec knowledge has any precursors in the
stream of literate cultures that came before.
Thus, we mainly get at ancient soil knowledge indirectly from archaeology
the different types and contexts of terraces and field walls, and from geoarchaeol-
ogy the biophysical, chemical and fossil evidence of remnant soils. These studies
have shown impressively diverse agricultural architecture, especially since the Classic
Periods, with many kinds of terracing and field walls that show extended control
of the rural landscape: terraces across channels, on slope benches, on foot-slopes,
on alluvial fans, on contours, and diversions toward and away from fields and away
from freshwater sources. Thus, research does show the long stream of indigenous
Mesoamerican soil knowledge, but we are just at the start of testing these terrace
soils for the full suite of proxies that might give us ideas about how ancient people
might have managed and altered soils for crop growth. Many papers have argued
that these soil beds were heavily fertilised from night soil, organic matter and spe-
cifically periphyton to get at the riddle of ancient subsistence; whereas some have
argued they were depleted by overuse to get at the riddle of collapse. Findings have
been controversial as we have refined acceptable methods, but many new studies
are coming on line that should begin to clarify our view of soil history.
Finally, the growing evidence for the impacts of pioneer farming parallel
findings in many parts of the world, such as the buried soils and post-settlement
alluvium of many parts of the United States (Beach 1994) and Mediterranean valleys
(Wilkinson 2005). This is a fairly obvious idea because of its universality, but there
is ample evidence that many development and natural resource planners from
79
Mesoamerica and the Caribbean Islands
References
Abrams, E. M. and D. J. Rue. 1988. The Causes and Consequences of Deforestation Among
the Prehistoric Maya. Human Ecology 16: 37795.
Atran, S. 2003. Anthropogenic Vegetation: A Garden Experiment in the Maya Lowlands.
In A. Gomez-Pompa, M. Allen, S. Fedick, J.Jimnez-Osornio (eds), Lowland Maya
Area: Three Millennia at the Human-Wildland Interface (Binghamton, NY: Haworth
Press), pp. 51732.
Atran, S., D. Medin, N. Ross, E. Lynch, J. Coley, E. Ucan Ek and V. Vapnarsky. 1996.
Folkecology and Commons Management in the Maya Lowlands. Proceedings of the
National Academy of Science, 96: 7598603.
Barba, L., B. Ludlow, L. Manzanilla and R. Valadez. 1987. La Vida Domstica De Teoti-
huacan: Un Estudio Interdisciplinario. Ciencia y Desarrollo 77: 2133.
Barba, L. and A. Ortiz. 1992. Anlisis Qumico de Pisos de Ocupacin: un Caso Etnogrfico
en Tlaxcala, Mexico. Latin American Antiquity, 3: 6382.
Barba, L., A. Ortiz, K. Link, L. Lpez Lujn and L. Lazos. 1996. Chemical Analysis of
Residues in Floors and the Reconstruction of Ritual Activities at the Templo Mayor,
Mexico. In M. V. Orna (ed.), Archaeolgical Chemistry: Organic, Inorganic, and Biochemical
Analysis (Washington, DC: American Chemical Society), pp.13956.
Barrera, M.A., A. Gmez-Pompa and C. Vzquez-Yez. 1977. El Manejo de las Selvas por
los Mayas. Bitica, 2(2):4760.
Beach, T. 1992. Estimating Erosion in Medium-Size Drainage Basins. Physical Geogra-
phy,13:20624.
Beach, T. 1994. The Fate of Eroded Soil: Sediment Sinks and Sediment Budgets of Agrarian
Landscapes in Southern Minnesota, 18511988. Annals of the Association of American
Geographers, 84:528.
Beach, T. 1998a. Soil Catenas, Tropical Deforestation, and Ancient and Contemporary Soil
Erosion in the Petn, Guatemala. Physical Geography, 19(5): 378405.
Beach, T. 1998b. Soil Constraints on Northwest Yucatn, Mexico: Pedoarchaeology and
Maya Subsistence at Chunchucmil. Geoarchaeology, 13(8):75991.
Beach, T. and N. Dunning. 1995. Ancient Maya Terracing and Modern Conservation in the
Petn Rain Forest of Guatemala. Journal of Soil and Water Conservation, 50: 13845.
Beach, T. and N. Dunning. 1997. An Ancient Maya Reservoir and Dam at Tamarindito,
Petn, Guatemala. Latin American Antiquity, 8(1):2029.
Beach, T., N. Dunning, S. Luzzadder-Beach and V. Scarborough. 2003a. Depression Soils
in the Lowland Tropics of Northwestern Belize: Anthropogenic and Natural Origins.
In A. Gomez-Pompa, M. Allen, S. Fedick and J. Jimnez-Osornio (eds), Lowland Maya
Area: Three Millennia at the Human-Wildland Interface (Binghamton, NY: Haworth
Press), pp.13974.
80
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Butzer, K.W. 1992b. From Columbus to Acosta: Science, Geography, and the New World.
Annals of the Association of American Geographers, 82: 54365.
Butzer, K.W. 1993. No Eden in the New World. Nature, 362: 1517.
Butzer, K.W. & E.K. Butzer. 1993. The Sixteenth-Century Environment of the Central
Mexican Bajo: Archival Reconstruction from Colonial Land Grants and the Question
of Spanish Ecological Impact. In K. Mathewson (ed.), Culture, Form, and Place: Essays
in Cultural and Historical Geography, Geoscience and Man, 32 (Baton Rouge: Louisiana
State University), pp.89124.
Caballero, J. 1992. Maya Homegardens: Past, Present, and Future. Ethnoecologica 1:
3553.
Carlos Donado, Luis. 1996. Characteristics and Potentials of Selected Soils in the Peten
Region of Guatemala. Unpublished Masters Thesis, Plant and Soil Science Department,
Knoxville: University of Tennessee.
Carter, W.E. 1969. New Lands and Old Traditions: Kekchi Cultivators in the Guatemalan
Lowlands. Gainsville: University of Florida Press.
Chase, A. and D. Chase. 1998. Scale and Intensity in Classic Period Maya Agriculture: Ter-
racing and Settlement at the Garden City of Caracol, Belize. Culture and Agriculture,
20(2/3): 6077.
Conservation Foundation. 1954. Soil Erosion Survey of Latin America, reprinted in the
Journal of Soil and Water Conservation, 9.
Conyers, L. and H Spetzler. 2002. Geophysical Exploration at Ceren. In P. Sheets (ed.),
Before the Volcano Erupted (Austin: University of Texas Press), pp2432.
Cook, D., B. Kovacevich, T. Beach and R. Bishop. 2006. Deciphering the Inorganic Geo-
chemical Record of Ancient Human Activity using ICP-MS: Two Examples from Late
Classic Soil Floors at Cancun, Guatemala. Journal of Archaeological Science, 33(5):
62840.
Cook, S.F. 1948. The Historical Demography and Ecology of the Teotlalpan, Ibero-Americana
33. Berkeley and Los Angeles: University of California Press.
Cook, S.F. 1949. Soil Erosion and Population in Central Mxico, Ibero-Americana 34. Berkeley
and Los Angeles: University of California Press.
Cooke, R.U. and R.W. Reeves. 1976. Arroyos and Environmental Change in the American
Southwest. Oxford: Clarendon Press.
Cordova, C. and J. Parsons. 1997. Geoarchaeology of an Aztec Dispersed Village. Geo
archaeology, 12(3): 177210.
Coultas, C.L., M.E. Collins and A.F. Chase. 1993. The Effects of Ancient Maya Agri-
culture on Terraced Soils of Caracol, Belize. In J.E. Foss, M.E. Thompson and M.W.
Morris (eds), Proceedings of the First International Pedo-Archaeology Conference, Special
Publication No. 9303 (Knoxville, TN: Agricultural Experiment Station, University of
Tennessee), pp.191210.
Coultas, C.L., Y.P. Hsieh and T.J. Post. 1997. A Toposequence of Soils in Northern Belize,
Central America. Soil and Crop Science of Florida, 56: 11724.
Cowgill, U.M. 1962. An Agricultural Study of the Southern Maya Lowlands. American
Anthroplogist, 64: 27386.
82
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Curet, L.A. 2003. Issues on the Diversity and Emergence of Middle-Range Societies of the
Ancient Caribbean: A Critique. Journal of Archaeological Research, 11: 142.
Dahlin, B., T. Beach, S. Luzzadder-Beach, D. Hixson, S. Hutson, A. Magnoni, D. Mansell
and D. Mazeau. In press. Reconstructing Agricultural Self-Sufficiency at Chunchucmil,
Yucatan, Mexico. Ancient Mesoamerica.
de Landa, Diego. 1982. Relacin de las Cosas de Yucatn. Mexico City: Editorial Porrua.
Denevan, W. 1992. The Pristine Myth: The Landscape of the Americas in 1492. Annals of
the Association of American Geographers, 82(3): 36985.
Diamond, Jared. 2004. Collapse: How Societies Choose to Fail or Succeed. New York:
Viking.
Donnelly, T.W., G. Horne, R. Finch and E. Lpez-Ramos. 1990. Northern Central America;
the Maya and Chortis Blocks. In G. Dengo and J. Chase (eds), The Caribbean Region:
The Geology of North America (Boulder, Colorado: Geological Society of America).
Donkin, R. 1979. Agricultural Terracing in the Aboriginal New World, Viking Fund Publica-
tions in Anthropology No. 56. Tucson, AZ:. University of Arizona Press.
Doolittle, W.E. 1985. The Use of Checkdams for Protecting Downstream Agricultural
Lands in the Prehistoric Southwest: A Contextualist Analysis. Journal of Anthropological
Research, 41: 279305.
Doolittle, W.E. 1990. Terrace Origins: Hypotheses and Research Strategies. Yearbook of
the Conference of Latin Americanists Geographers, 16: 947.
Doolittle, W.E. 1995. Indigenous Development of Mesoamerican Irrigation. Geographical
Review, 85: 30123.
Doolittle, W.E. 2000. Cultivated Landscapes of Native North America. Oxford: Oxford
University Press.
Dunning, N.P. 1992. Lords of the Hills: Ancient Maya Settlement in the Puuc Region, Yucatn,
Mexico, Monographs in World Prehistory 15. Madison, WI: Prehistory Press.
Dunning, N.P. and T. Beach. 1994. Soil Erosion, Slope Management, and Ancient Terracing
in the Maya Lowlands. Latin American Antiquity, 5(1): 5169.
Dunning, N.P. and T. Beach. 2000. Stability and Instability in Pre-Hispanic Maya Land-
scapes. In D. Lentz (ed.), Imperfect Balance: Landscape Transformations in the Precolumbian
Americas (New York: Columbia University Press), pp. 179202.
Dunning, N.P. and T. Beach. 2004a. Fruit of the Luum: Lowland Maya Soil Knowledge
and Agricultural Practices. Mono y Conejo: Journal of the Mesoamerican Archaeological
Research Laboratory, 2: 315.
Dunning, N.P. and T. Beach. 2004b. Noxious or Nurturing Nature? Maya Civilization
in Environmental Context. In C. Golden and G. Borgstede (eds), Continuities and
Changes in Maya Archaeology: Perspectives at the Millennium (New York: Routledge),
pp.12541.
Dunning, N.P., T. Beach and D. Rue. 1997. The Paleoecology and Ancient Settlement of
the Petexbatn Region, Guatemala. Ancient Mesoamerica, 8: 112.
Dunning, N.P., T. Beach, P. Farrell and S. Luzzadder-Beach. 1998. Prehispanic Agricultural
Adaptive Regions in the Maya Lowlands. Culture and Agriculture, 20(4): 87101.
83
Mesoamerica and the Caribbean Islands
Dunning, N.P., S. Luzzadder-Beach, T. Beach, John Jones, V. Scarborough and T.P. Culbert.
2002. Arising from the Bajos: The Evolution of a Neotropical Landscape and the Rise of
Maya Civilization. Annals of the Association of American Geographers, 92(2): 26783.
Eidt, R.C. 1984. Theoretical and Practical Considerations in the Analysis of Anthrosols.
In G. Rapp, Jr., G. Gifford and J.A. Gifford (eds), Archaeological Geology (New Haven:
Yale University Press), pp. 15590.
Evans, S.T. 1992. The Productivity of Maguey Terrace Agriculture in Central Mexico
during the Aztec Period. In T.W. Killion (ed.), Gardens of Prehistory: The Archaeology
of Settlement Agriculture in Greater Mesoamerica (Tuscaloosa, Alabama: University of
Alabama Press), pp. 92115.
Food and Agriculture Organization of the United Nations (FAO). 2005. The State of the
Worlds Forests. Rome: United Nations.
Fedick, S.L.1994. Ancient Maya Agricultural Terracing in the Upper Belize River Area.
Ancient Mesoamerica, 5: 10727.
Fedick, S.L. 1996. An Interpretive Kaleidoscope: Alternative Perspectives on Ancient Agri-
cultural Landscapes in the Maya Lowlands. In S.L. Fedick (ed.), The Managed Mosaic:
Ancient Maya Agriculture and Resource Use (Salt Lake City: University of Utah Press),
pp.10731.
Fedick, S.L. 2003. Archaeological Evidence for Ancient and Historic Resource Use Associated
with the El Edn Wetland, Northern Quintana Roo, Mexico. In A. Gomez-Pompa, M.
Allen, S. Fedick and J.Jimnez-Osornio (eds), Lowland Maya Area: Three Millennia at
the Human-Wildland Interface (Binghamton, NY: Haworth Press), pp.33960.
Fedick, S.L., B.A. Morrison, B.J. Andersen, S. Boucher, J.C. Acosta and J.P. Mathews. 2000.
Wetland Manipulation in the Yalahau Region of the Northern Maya Lowlands. Journal
of Field Archaecology, 27(2): 13152.
Fernndez, F.G., R. Terry, T. Inomata and M. Eberl. 2002. An Ethnoarchaeological Study
of Chemical Residues in the Floors and Soils of Qeqchi Maya Houses at Las Pozas,
Guatemala. Geoarchaeology, 17: 487519.
Fernandez, F., K. Johnson, R. Terry, S. Nelson and D.Webster. 2005 Soil Resources of the
Ancient Maya at Piedras Negras, Guatemala. Soil Science Society of America Journal 69:
202032.
Fisher, C., H.P. Pollard, I. Israde, V.H. Garduno and S. Banerjee. 2003. Reexamination of
Human Induced Environmental Change Within the Lake Ptzcuaro Basin, Michoacn,
Mexico. Proceedings of the National Academy of Sciences,100(8): 495762.
Folan, W.J., L.A. Fletcher and E.R. Kintz. 1979. Fruit, Fiber, Bark, and Resin: The Social
Organization of a Maya Urban Center, Coba, Quintana Roo, Mexico. Science, 204:
697701.
Furley, P.A. 1987. Impact of Forest Clearance on the Soils of Tropical Cone Karst. Earth
Surface Processes and Landforms, 12: 5239.
Garcia-Oliva, F., L. Martinez and J.M. Maass. 1995. Long-Term Net Soil Erosion as De-
termined by 137Cs Redistribution in an Undisturbed and Perturbed Tropical Deciduous
Forest Ecosystem. Geoderma, 68: 13547.
84
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Reid, L. and Dunne, T. 1996. Rapid Evaluation of Sediment Budgets. Cremlingen, Germany:
Catena Verlag Press.
Rice, D.S. and T.P. Culbert. 1990. Historical Contexts for Population Reconstruction in the
Maya Lowlands. In T.P. Culbert and D.S. Rice (eds), Precolumbian Population History
in the Maya Lowlands (Albuquerque: University of New Mexico Press), pp. 136.
Rincon Maunter, C. 1999. Man and the Environment in the Coixtlahuaca Basin of North-
western Oaxaca, Mexico: Two Thousand Years of Historical Ecology. Unpublished Ph.
D. Dissertation, Austin: University of Texas.
Robinson, E.J., P. Farrell, K.F. Emery and G.E. Braswell. 2002. Preclassic Settlements and
Geomorphology in the Highlands of Guatemala: Excavations at Urias, Valley of Anti-
gua. In M. Love, H. Escobedo and M. Popenoe de Hatch (eds), Incidents of Archaeology
in Central America and Yucatn (Lanham, MD: University Press of the Americas), pp.
25176.
Robles Ramos, R.1950. Apuntes Sobre la Morfologia de Yucatn. Boletn de la Sociedad
Mexicana de Geografa y Estadstica, 69: 27106.
Rosenmeier, M., D. Hodell, M. Brenner, J. Curtis and T. Guilderson. 2002. A 4,000-Year
Lacustrine Record of Environmental Change in the Southern Maya Lowlands, Petn,
Guatemala. Quaternary Research 57: 18390.
Rouse, I. 1992. The Tainos: Rise and Fall of the People Who Greeted Columbus. New Haven,
CT: Yale University Press.
Sahagn, Fray Bernardino de. 1900 [1577]. General History of the Things of New Spain [1577],
2nd revised edn, 4 vols. Santa Fe, New Mexico: School of American Research.
Sauer, C.O. 1952. Agricultural Origins and Dispersals. New York: American Geographical
Society.
Sauer, C.O. 1966. The Early Spanish Main. Berkeley: University of California Press.
Scarborough, V. 1993. Water Management in the Southern Maya Lowlands: An Accretive
Model for the Engineered Landscape. In V. Scarborough and B. Isaac (eds), Economic
Aspects of Water Management in the Prehispanic New World, Research in Economic An-
thropology Supplement 7 (Greenwich, CT: JAI Press), pp. 1768.
Shang, C. and H. Tiessen. 2003. Soil Organic Sequestration and Stabilization in Karstic
soils of the Yucatn. Biogeochemistry, 62: 17796.
Sheets, Payson (ed.). 2002. Before the Volcano Erupted: The Ancient Ceren Village in Central
America. Austin: University of Texas Press.
Sluyter, A. 1994. Intensive Wetland Agriculture in Mesoamerica: Space, Time and Form.
Annals of the Association of American Geographers, 84(4): 55784.
Sluyter, A. 1997. Regional Holocene Records of the Human Dimension of Global Change:
Sea-Level and Land-Use Change in Prehistoric Mexico. Global and Planetary Change,
14: 12746.
Sluyter, A. and A.H. Siemens. 1992. Vestiges of Prehispanic, Sloping Field-Terraces on the
Piedmont of Central Veracruz, Mexico. Latin American Antquity, 3: 14860.
Smith, M.E. and T.J. Price. 1994. Aztec-Period Agricultural Terraces in Morelos, Mexico:
Evidence for Household-level Agricultural Intensification. Journal of Field Archaeology,
21: 16979.
89
Mesoamerica and the Caribbean Islands
Smyth, M.P., C.D. Dore and N. Dunning. 1995. Interpreting Prehistoric Settlement Pat-
terns: Lessons from the Maya Center of Sayil, Yucatn. Journal of Field Archaeology,
22: 32146.
Sommer, M. and E. Schlicting. 1997. Archetypes of Catenas in Respect to Matter A
Concept for Structuring and Grouping Catenas. Geoderma 76: 133.
Spores, R. 1969. Settlement, Farming Technology, and Environment in the Nochixtlan
Valley. Science, 166: 55769.
Steggerda, M. 1941. Maya Indians of Yucatn, Publication 531. Washington, D.C.: Carnegie
Institution of Washington.
Stevens, R.L. 1964. The Soils of Middle America and their Relation to Indian Peoples and
Cultures. In R.C. West (ed.), Natural Environment and Early Cultures, Handbook of
Middle American Indians, (Austin: University of Texas Press), vol. 1, pp. 265315.
Tabo, E.E. 1989. Arqueologa: Agricultura Aborigin Antillana. Havana: Editorial de Ciencias
Sociales.
Terry, R.E., P. Hardin, S. Houston, M.W. Jackson, S.D. Nelson, J. Carr and J. Parnell. 2000.
Quantitative Phosphorus Measurement: A Field Test Procedure for Archaeological Site
Analysis At Piedras Negras, Guatemala. Geoarchaeology, 15: 15166.
Thompson, J.E.S. 1954. The Rise and Fall of Maya Civilisation. Norman: University of
Oklahoma Press.
Treacy, J.M. and W. Denevan. 1994. The Creation of Cultivable Land Through Terracing.
In N. Miller and K. Gleason (eds), The Archaeology of Garden and Field (Philadelphia:
The University of Pennsylvania Press), pp. 91110.
Turner, B.L. II. 1974. Prehistoric Intensive Agriculture in the Mayan Lowlands. Science,
185(4146): 11824.
Turner, B.L. II. 1976. Prehistoric Population Density in the Maya Lowlands: New Evidence
from Old Approaches. Geographical Review, 66: 7382.
Turner, B.L. II. 1979. Prehispanic Maya Terracing in the Central Maya Lowlands: Problems
of Agricultural Intensification. In N. Hammonds and G.R. Willey (eds), Maya Archaeol-
ogy and Ethnohistory (Austin: Univ. of Texas Press), pp. 10315.
Turner, B.L. II. 1983. Once Beneath the Forest: Prehistoric Terracing in the Rio Bec Region of
the Maya Lowlands. Boulder, CO.: Westview.
Tuan, Y., 1966. New Mexican Gullies: A Critical Review and Some Recent Observations.
Annals of the Association of American Geographers, 56: 57397.
Vaughan, H.H., E. Deevey and S. Garrett-Jones. 1985. Pollen Stratigraphy of Two Cores
from the Petn Lake District, with an Appendix on Two Deep-water Cores. In M. Pohl
(ed.), Prehistoric Lowland Maya Environment and Subsistence Economy (Cambridge, MA:
Harvard University Press), pp. 7389.
Veloz Maggiolo, M. 1976. Mediombiente y Adaptacin Humana en la Prehistoria de Santa Do-
mingo, Coleccin Historia y Sociedad 24. Universidad Autnoma de Santo Domingo.
Wahl, D., R. Byrne, T. Schreiner and R. Hansen. In Press. Holocene Vegetation Change in
the Northern Petn and Implications for Maya Prehistory. Quaternary Research.
90
Tim Beach, Sheryl Luzzadder-Beach and Nicholas Dunning
Webb, E., H. Schwarz and P. Healy. 2004. Carbon Isotope Evidence for Ancient Maize
Agriculture in the Maya Lowlands. Journal of Archaeological Science, 31: 103952.
Wells, E.C., R Terry, J. Parnell, P. Hardin, M. Jackson and S. Houston. 2000. Chemical
Analyses of Ancient Anthrosols in Residential Areas at Piedras Negras, Guatemala.
Journal of Archaeological Science, 27: 44962.
Whitmore, T. and B. Turner II. 2001. Cultivated Landscapes in Mesoamerica on the Eve of
the Conquest. Oxford: Oxford University Press.
Wilken, G. 1987. Good Farmers:Traditional Agricultural Resource Management in Mexico and
Central America. Berkeley: University of California Press.
Wilkinson, T. 2005. Soil Erosion and Valley Fills in the Yemen Highlands and Southern
Turkey: Integrating Settlement, Geoarchaeology, and Climate Change. Geoarchaeology,
20(2): 16992.
Williams, B. 1972. Tepetate in the Valley of Mexico. Annals of the Association of American
Geographers, 62: 61826.
Williams, B. 1989. Contact Period Rural Overpopulation in the Basin of Mexico: Car-
rying Capacity Models Tested with Documentary Data. American Antiquity, 54(4):
71532.
Williams, B. and H.R. Harvey. 1997. The Cdice de Santa Mara Asuncin: Facsimile and
Commentary: Households and Lands in Sixteenth-Century Tepetlaoztoc. Salt Lake City:
University of Utah Press.
Williams, B. and C.A. Ortiz-Solorio. 1981. Middle American Folk Soil Taxonomy. Annals
of the Association of American Geographers, 71(3): 33558.
Williams, L. 1990. Agriculture Terrace Evolution in Latin America. Yearbook of the Confer-
ence of Latin Americanist Geographers, 16: 8293.
Williams, L. and B. Walter. 1988. Controlled-erosion Terraces in Venezuela. In W. Mold-
enhauer and N.Hudson (eds), Conservation Farming on Steep Lands. Ankeny, Iowa: Soil
and Water Conservation Society of America.
Wilson, S.M., H. Iceland and T. Hester. 1998. Preceramic Connections Between Yucatn
and the Caribbean. Latin American Antiquity, 9: 34252.
Wiseman, F. 1983. Analysis of Pollen from the Fields at Pulltrouser Swamp. In B.L. Turner
and P.L. Harrison (eds), Pulltrouser Swamp: Ancient Maya Habitat, Agriculture, and Set-
tlement in Northern Belize (Austin: University of Texas Press), pp. 10519.
Wingard, J. 1996. Interactions Between Demographic Processes and Soil Resources in
the Copn Valley, Honduras. In S. Fedick (ed.), The Managed Mosaic: Ancient Maya
Agriculture and Resource Use (Salt Lake City: The University of Utah Press).
Woods, W.I. and J. M.McCann. 1999. The Anthropogenic Origin and Persistence of
Amazonian Dark Earths. The Yearbook of the Conference of Latin American Geographers,
25: 714.
91
Wetlands in the Americas
Standing in stark contrast to drylands are wetlands. Whereas the former gener-
ally suffer from a paucity of moisture that inhibits agriculture, the latter have an
abundance that also places extreme limits on cultivation. There are several types
of wetlands ... they can all be considered areas transitional between terrestrial and
aquatic ecosystems where the water table is usually at or near the surface, or land is
covered by shallow water and where at least one of the three following conditions
exists: the land periodically, if not always, supports predominantly hydrophytes; the
substrate is undrained hydric soil; or the substrate is saturated non-soil (Doolittle
2000, p.413).
Introduction
One of the most important chapters of Pre-Columbian soil history is Native
American manipulation of wetland soils. Many wetlands in the Americas are so
altered by humans that it is often difficult to draw a line between what is cultural
and what is natural (Denevan 1992, Erickson 2000, p.317). One reason why is
that human interaction with wetlands in the Americas stretches far back to the
early migrations from Beringia. Wetlands often preserved the earliest remains of
culture in artefacts and human remains (Purdy 2001), including at one of the earliest
sites at Monte Verde (Meltzer 1997). Wetlands sediments also preserve ecological
records of past environments and human activity in the proxy records of paleosols,
pollen, mollusc shells and other organic remains, and oxygen isotopes. But some
wetlands themselves may be considered cultural artefacts, either unintentionally
or meticulously planned. Two examples are deforestation, which through reduced
transpiration can induce water tables to rise and wetlands to form, and wetlands
that have been ditched and reformed into wetland fields by ancient civilisations
92
Sheryl Luzzadder-Beach and Tim Beach
all over the Americas. This latter interaction closely links the history of soils to
hydrology and culture and represents one of the most widespread and intensive
landscape changes by Pre-Columbian people.
Native Americans transformed wetlands of many types into a variety of
patterns, including ridged fields, ditched fields, sunken gardens, and chinampas
or so-called floating gardens. To find a baseline or pristine environment for
comparison with wetland alternation in the Americas, we need to look as far back
as the end of the Pleistocene (Denevan 1992; Erickson 2000). Since this history
is so long and social and biophysical factors are so complex, research on wetland
complexes has used an extensive tool-kit from a broad cross section of disciplines.
Some disciplines have been predisposed to a biophysical perspective that natural
processes are ubiquitous and dominant in soil formation, and other disciplines are
predisposed to a cultural perspective; wherein:
The dominant perspective is that cultural systems encapsulate, mediate, and
translate natural systems so thoroughly that the properties of the physical world
are, in essence, irrelevant to our understanding of human societies ... Culture is
capable of complete intermediation of the natural world (Kolata 2000, p.163).
The examples in this chapter show that physical factors are also important to culture
and a relevant part of the human history of soils.
Many social scientists continue to be wary of their long history with
nineteenth-century geographical and environmental determinism (Kolata 2000,
p.164), and focus only on culture within the theory of mutual embeddedness of
the material and the cultural (Kolata 2000, p.163). But over the last few decades,
growing scientific research has repeatedly demonstrated that human action and the
environment are complex, coupled systems. Scholars have used historical, geological,
archaeological, ecological, and particularly geographical perspectives to interpret
anthropogenic wetland environments. To these ends, many have reviewed progress
in this field in indigenous Mesoamerica (see Chapter 4, also Sluyter 1994; Whitmore
and Turner 2001); and reviewed progress and these approaches for ancient South
America (e.g., Denevan 2002; Erickson 2000). Denevan (1992) and Doolittle
(1992, 2000) summarised much of what is known about aboriginal agriculture in
North America. Herein, we follow the threads of the major evidence about wetland
soils manipulation in the Americas.
Pioneering work discovering and interpreting anthropogenic wetland land-
scapes in Pre-Columbian Latin America began largely with the efforts of geographers
and archaeologists in the late 1960s and early 1970s (Parsons and Bowen 1966;
Denevan 1970; Armillas 1971; Turner and Harrison 1983; Sluyter 1994). This
work had it roots in expeditions like the Carnegie Institutions An Archaeological
Reconnaissance by Air in Central America in 1929, piloted by Col. Charles A.
93
Wetlands in the Americas
Lindbergh and published in The Geographical Review (Ricketson and Kidder 1930,
p.177). This research was an early venture in remote sensing and surveyed ancient
Maya sites and environments in Guatemala, Belize and the Yucatn Peninsula (Figure
5.1). Other early twentieth-century research efforts, like the Carnegies own work
in the Yucatn (Steggerda 1941), studied the broad resource base of Maya sites, but
many mid-century studies were focused on specific archaeology sites (Dunning and
Beach 2004). As archaeological studies expanded to broader regional surveys in the
1960s and 1970s, they found more and more wetland patterns based on increas-
ingly sophisticated aerial photography. For more than 40 years, since Palerm and
Wolf (1957, p.28) first hypothesised the ground patterns as chinampas, scholars
of several disciplines have worked to understand these patterns (Coe 1964; Parsons
1976; Nichols and Frederick 1993), but there is still much debate about how they
formed in the changing environments of the Late Holocene.
Modern excavations of wetland fields include Armillas (1971) groundbreak-
ing research in the basin of Mexico, but because wetlands are difficult to excavate
and their cultural artefacts unpromising, much early work on wetland fields in the
Americas was on mapping patterns. Pioneering research by Turner and Harrison
(1981, 1983) in Pulltrouser Swamp, Belize, began an era of more detailed stud-
ies of these remarkably patterned sites. The most frequent interpretation of the
wetland sites is that humans created or modified them for agriculture. A variety of
classifications of wetland fields shapes and functions grew from this field work, but
another interpretation was that they are patterned ground formed by a combination
of natural and anthropogenic processes (Jacob 1995; Pope et al. 1996; Beach and
Luzzadder-Beach 2002; Luzzadder-Beach et al. 2003). Turner and Harrison con-
cluded that their particular wetland features in Pulltrouser Swamp were not natural
features such as gilgai (patterned ground formed from vertisols under alternating
wet and dry conditions). Continued fieldwork based on excavation and reconstruc-
tion has revealed more detail about the soil types, crops and ecology of these sites
across the Americas (Turner and Harrison 1983; Denevan 2002; Erickson 2000;
Kolata 2000; McClung de Tapia 2000). In this chapter we focus on significant
Pre-Columbian sites in the Americas, including a typology of wetland soils and
patterns, with regional discussions of the wetlands of Central Mexico; wetland and
bajo agriculture in the Maya Lowlands of Mesoamerica; South America, regionally
divided between the Andean Highlands and seasonally flooded lowlands in South
America; and floodplain agriculture in North America (Figure 5.1).
Wetland Soils
Wetland soils in general are unique in that the climate component of soil formation
is less important than in upland soils. Wetland soils are soils influenced by the sea-
sonal or perennial influence of water. These soils vary according to how much of the
95
Wetlands in the Americas
time water inundates or floods them. Soils that have seasonal inundation tend to be
mineral soils with mottled colours: greys from water-saturated reducing conditions,
and yellows or oranges indicative of seasonally well-drained, oxidising conditions.
Soil scientists give the horizons (e.g. B or C) of these mineral soils a g designator
for the gleyed-reduced minerals of the soil. Soils that have a constantly high water
table are dominated by reduction and tend to be grey in colour, or sometimes pale
blue and green. Soils that are inundated by water most of the time are dominated
by higher fractions of organic matter, and thus are darker in colour, often dark
browns and black. One soil order, histosol, is waterlogged and thus dominated by
organic matter for more than half of the upper 80cm (Soil Survey Staff 1998).
Common terms for histosols are peats and mucks, depending on whether they have
more or less organic matter. Mineral soils that are waterlogged fall into Aquic or
other Great Groups of such soil orders as entisols, vertisols, inceptisols, mollisols
and others. These soil characteristics have many implications for crops and natural
vegetation communities that can grow in these conditions. Different crops have
different requirements at different times in their growing cycles for moisture, but
most grow best with plentiful micropore water held electrostatically to soil surfaces,
and little macropore water that would reduce aeration of plant roots and aerobic
microbial communities. This explains the imperative to drain inundated soils and
build up (or raise) fields far enough above the water table for root growth to take
advantage of micropore water and aeration in macropores.
in the rainy season, and delivering irrigation water during drier periods. William
Denevan, one of the key scholars of Pre-Columbian landscape manipulation, has
synthesised three main reasons for canalising wetlands in South America. The overall
function of ditching fields was to expand soil resources (Denevan 2002, p.264) by
building anthrosols mounds. A secondary function was for irrigation or drainage,
but further investigations of sites near Lake Titicaca, for example, revealed that
ditch or depression patterns in many field systems did not seem to function to
carry water away from the sites, but rather, may have been built to block flooding
(Denevan 2002, p.260). Third, the typical layout of ditches and fields may be for
energy conservation, using waters high specific heat to help provide night-time frost
protection in highland sites, especially depression areas prone to cold air drainage
(Erickson 2000; Denevan 2002, p.266).
Extensive excavation and mapping have discerned many other purposes of
these raised fields beyond the basic three of reclamation, irrigation-drainage and
warming. Raised field agriculture manages soil water by controlling water tables
to maintain aeration, drainage and soil moisture infiltration. Soils manipulation
also supplies and recycles nutrients through transferring organic muck and accu-
mulated sediments from the adjacent canals to the planting platforms (Erickson
2000, p.334). Wetland manipulations also provide an integrative agriculture and
aquaculture system (akin to paddy rice ecosystems) through expanded and mobile
wetland canal and dam management (Erickson 2000). Another function is to
control water quality. Ancient farmers could have managed the canal and dam
systems to improve water quality, separating fresh water from saline and alkaline
waters (Denevan 2002, p.269, citing Erickson 1992).
Several studies have gone beyond excavation and mapping to understand
wetland fields by reconstructing field systems, employing modern scientific agricul-
tural research, and poring through ethnohistorical literature. In Peru and Bolivia,
Erickson used all of these approaches, testing these functions through experimental
rehabilitation of ancient wetland fields (Erickson 2000; Denevan 2002) and finding
precedence in ethnographic studies and validation in modern agricultural research
(Erickson 2000). In Tabasco and Quintana Roo, Mexico, Gliessman (1991) studied
both ancient and modern indigenous agriculture on raised field sites. He found
modern fields farmed using indigenous methods to be an average of four times more
productive than fields prepared and farmed mechanically. The abundant yields of
wetland raised fields form one of the keys to understanding the population levels,
urbanisation and civilisation of several regions of Pre-Columbian America. There
have been other successful chinampas rehabilitation projects elsewhere in Mesoa-
merica, but they have been far from panaceas (Gomez-Pompa 1982). One was a
state-run project by Tabasco Mexico in the 1970s, called the Camelons Chontales
Project. Chapin (1988) explains how they tried to build chinampas but made many
mistakes reminiscent of the mistakes made by early wetland restoration projects in
97
Wetlands in the Americas
the United States (National Research Council 1992). The failure of this project does
not reduce the potential or past importance of wetland agriculture, but exemplifies
typical errors of agricultural development not including local people, taking ac-
count of ancient knowledge, or precisely planning the soils and hydrology. Indeed,
after the government gave up on the Chontales Project, a local Maya group made
the fields productive (Erickson 1998, p.42).
Central Mexico
with deliberately planted trees such alders (Alnus) and willows (Salix) (Parsons and
Denevan 1989[1967], p.219).
The native alders and willows stabilise the island margins, and ancient
farmers layered strips of aquatic vegetation across the islands, building up a vegeta-
tive net to hold and raise the soil bed. Indigenous farmers cleaned the canals and
maintained island fertility by scooping lake-bottom mud and organic matter onto
these vegetative mats to build up the artificial islands (Coe 1964; Armillas 1971).
These island fields ranged from about 3m wide and 30m long, up to squares of
100m or more per side (Parsons and Denevan 1989[1967]; Carrasco and Sessions
1998; Smith 2003). Crop irrigation occurred by the high water table, evident in
canal surfaces, capillary rise and water applied by hand from the canals. Since Lake
Texcoco is saline, farmers must guard against saltwater intrusion with a system of
dikes (Denevan 1970; Coe 1964; Armillas 1971).
The population of the Mexica capital of Tenochtitlan was extremely high,
estimated from 200,000 to at 300,000 in Aztec times (Sluyter 1994; Carrasco and
Sessions 1998), and required remarkably intensive food production systems such
as upland terraces and the chinampas. The chinampas used a fine-tuned rotation
and multi-cropping of agricultural commodities (Denevan 1970). Sluyter used an
estimate of 9,000ha of chinampas and their yield ranges to estimate food production
enough to sustain a population of 171,000 annually (Sluyter 1994). The types of
crops that could have been grown on these plots included amaranth, squash, maize,
beans, chili and flowers among others, rotated in from seed beds or floating seed
mats (West and Augelli 1966; Carrasco and Sessions 1998; Smith 2003).
Central America
(Turner and Harrison 1983; Siemens and Puleston 1972; Siemens 1983; Siemens
et al. 1988; Jacob 1995; Jacob and Hallmark 1996; Pohl et al. 1996).
The lines of evidence that bear upon these questions are surface patterns
stratigraphically uncovered from excavations, soil and water characteristics, pale-
oecological proxies, and artefacts. Conclusions from any one area may not have
regional implications, because wetlands are a complex of several finely tuned pa-
rameters of an environment that might change dramatically with slight changes in
any one parameter. As in other regions of the Americas, polygonal surface patterns
occur all over the perennial wetlands of the Maya Lowlands, and many appear quite
strikingly as anthropomorphic and others as natural. In fact, some of these features
could be any number of possible natural features, and to identify canals reliably
requires mapping showing regional connections and careful excavation through the
full sequence of sediments. Adding further to the problem is modern disturbance.
Farmers are draining, burning and ploughing these areas, which is greatly altering
soil stratigraphy and organic preservation.
There exist several major views in the literature about the patterns. The first
research on the topic was careful to examine natural explanations such as gilgai, an
Australian aboriginal term for mounds formed from expanding and contracting
clays (Siemens and Puleston 1972). Much of this region has such clays, and the
gilgai topography occurs all over the bajos or seasonally wet environments, but most
gilgai are low enough for ploughing to remove them (Eswaran et al. 1999), though
they may still impart curious surface patterns when viewed from aerial photographs.
Some rectilinear features may indeed be gilgai, but gilgai could not be present in
truly perennially wet areas, because they required drying for clay contraction and
wetting for clay expansion to form features. Paleogilgai, though, could be inherited
in perennial wetlands from drier periods.
The earliest discovered fields were those on the coastal plain margins of the
Yucatn Peninsula (Figure 5.1, and Figure 4.2 in Chapter 4) (Siemens and Pule-
ston 1972) and next in coastal Veracruz, Mexico in the Olmec and Totonac areas
(Schmidt 1977; Siemens 1982). Siemens et al. (1988) studied wetland agriculture
in Yucatn, Mexico by excavating and using a series of cores to obtain samples
for analysing phytoliths, pollen and ceramics from different depths. They found
cross-shaped maize phytoliths comprised 1316 per cent of the fine silt fraction
at 90110cm depths and maize pollen reached 7 per cent of all pollen at 150cm.
The grey, silty-clay layer with most of the ceramics, economic pollen, and phyto-
liths also had plentiful charcoal and no layering, whereas the overlying sandy layers
had layering and few economic proxies and artefacts (Hebda et al. 1991; Siemens
1998). The dating based on ceramics could distinguish only that below 60cm the
sediments had Preclassic or Early Classic (Totonac) elements but the radiocarbon
dates were curiously modern (Siemens 1998).
100
Sheryl Luzzadder-Beach and Tim Beach
The major area of debate has been between the interpretations of Turner
and Harrisons projects (1981, 1983, and Harrison 1996) and of Bloom, Pohl and
Popes projects (1983, 1990 and 1996). Turner and Harrison used archaeological
site analysis together with surface mapping of rectilinear features and excavations
across the features to show the presence of wetland agriculture with both drained
and raised fields during the Late Classic. The main difference in interpretation of
stratigraphy is substantial. The Turner group interpreted two types of field: chan-
nelised fields at the dryland edge of wetlands and raised fields deeper in the wetlands,
which they reasoned Maya farmers built up from canal excavations. They focused
their interpretation on dominance of Late Classic artefacts in the upper soils of
these rectilinear features and reasoned that the underlying grey, mottled clay was
the Maya fill used to build fields. They used the following evidence to interpret
these as Late Classic planting surfaces: the geometric regularity of fields; the mot-
tled fill; the buried soil under the fill; artefacts occurring mainly above and below
the fill layer; and the fills pollen, indicative of a mixture of upland and wetland
species (Turner and Harrison 1983). Unfortunately the pollen had suffered much
degradation and the notion that this pollen indicated economic activity is less
secure (Turner and Harrison 1983, p.114).
The Pohl, Pope, and Bloom (Bloom et al. 1983; Pohl 1990; Pope et al.
1996; Jacob 1995; Pohl et al. 1996) group interpreted their many excavations
across similar features and a large excavation through one of Turners excavations
to show a different pattern (Pope et al. 1996, p.167). They substantially deepened
the excavation to obtain a longer chronology and clearer picture of the stratigraphy.
Based on thin section analysis of minerals and chemical analyses, they interpreted
the grey clay fill as a combination of naturally precipitated gypsum and carbonate
and anthropogenically eroded and then deposited clay (Pope et al. 1996, p.170).
Both groups found the following pattern: an upper organic clay soil that is 1050
cm thick, a sequence of sediments dominated by gypsum and calcium carbonate
that is about 100cm thick, and a variable paleosol (Pope et al. 1996, p.167; Turner
1983, p.45). Turners group found the paleosol only in auger cores, and inter-
preted it to be the old wetland soil upon which the Maya built the fields, whereas
Pope et al.s (1996, p.170) deeper pit provides evidence that it formed before the
Preclassic (based on ceramics and radiocarbon dates), had tremendous amounts
of maize pollen and charcoal from Preclassic Maya agriculture, and is similar to
the current swamp-forest soil of today. Farther into the wetland, the paleosol is
capped by 124cm thick peat that could only form near the surface, thus implying
the water table had risen. The upper peat of the wetter areas corresponds to the
gypsum-rich grey clay that started to form farther inland from about the Preclassic
to Classic Period boundary (c. ad200). Pohls group argued that the gypsum-rich
layers formed quickly because there is very little organic matter, or that the high
levels of sulphide catalysed organic matter oxidation. They explained that this grey
101
Wetlands in the Americas
clay formed as sea level rose and the mottled appearance was due to a mixture of
gypsum, carbonate and organic clay shown in the thin sections.
Evidence for Maya canal building may show different epochs responding
to different regional environmental impacts. Jacob (1995) studied ancient wetland
fields and canals at the site of Colha in Cobweb Swamp, Belize (Figure 4.1), reason-
ing that Maya canal building by the Classic or Preclassic was intended to manage
rising water levels. Pohl et al. (1996) confirmed Bloom et al.s findings, placing
the intensification of ditching in the Maya Lowlands at 1000bc, and abandon-
ment of the ditched wetland fields in Belize largely by the Classic Period due to
permanent flooding. Pohl et al. further reported that [f ]ield manipulations often
involved minor modifications of natural hummocks (Pohl et al. 1996, p.355).
These authors introduced the idea that the patterned or rectilinear fields may be
more natural in origin than initially thought, and that [c]anal systems are not as
extensive in Northern Belize as previously reported (Pohl et al. 1996, pp.355,
367). Pohl et al. (1996) limited these initial findings to low-lying areas subject to
sea level rise, and separated them from findings for highland areas.
Turners team provided foundational evidence for these complex human-
disturbed wetlands. Pohls team refined and refocused research on the paleosol and
the grey clay above it. They described a cascade of natural processes that explained
the soil sequences here: sea level rise drove up sulphate-rich water tables which
precipitated gypsum crystals into and above the Preclassic soil surface. All of this
buried this Preclassic landscape and its old soil surface. Both groups mentioned the
deep paleosol, but Pohls group recognised its plentiful charcoal, some artefacts and
that it had to form during times of much lower water tables, otherwise it would
have been a lacustrine (lake deposit) marl formed well below the current water ta-
ble. Turners team provided an appropriately broad chronology for the upper fields
based on ceramics: Late Preclassic to Terminal Classic (Turner and Harrison 1983,
p.254). Pohls group provided a longer chronology, showing how water inundation
massively altered this environment before or at the start of the Classic Period.
The rectilinear surface features formed in dynamic ecosystems were influ-
enced both by natural and Pre-Columbian human impacts. Human-induced change
resulted from increasingly widespread population growth and deforestation begin-
ning in the Preclassic (1200bcad250). This deforestation led to soil erosion that
depleted upland soils and aggraded lowlands, thus altering depressions and wetlands
from the Preclassic through the Late Classic (Beach 1998; Beach et al. 2002, 2003
and 2006 in press; Dunning et al. 2002; Hansen et al. 2002). Throughout this
period, several lines of evidence indicate that the Maya Lowlands were becoming
drier and that sea level was rising (Hodell et al. 1995 and 2000; Pohl et al. 1996;
High 1975). Droughts here have been linked to solar cycles (Hodell et al. 2001),
but deforestation, which could lead initially to increased runoff and higher water
tables through lower transpiration, has also been linked to longer term drying (Ray
102
Sheryl Luzzadder-Beach and Tim Beach
et al. 2005). Similar findings about deforestation and water table rise have been
reported in wetland sites following deforestation elsewhere in the world, including
Ireland after elm declines (Moore 1975).
Also during this period of environmental change and population growth, the
ancient Maya developed complex and diverse agricultural systems that conserved
upland soil and water and reclaimed lowland fields. Yet we have scant evidence
for intensive wetland agriculture in perennial wetlands of the core of the Maya
Lowlands. Sluyter (1994) points out that evidence within the Maya core exists
for only 15ha near Rio Azul, 1.5ha near Cerros and 375ha near Edzna, whereas
evidence for the periphery in Northern Belize, Rio Candelario and Bajo Morocoy
is much more extensive and may have persisted through the Classic Period. Since
the Maya had no beasts of burden, long distance food transport from these fields
was not economical, except for high-value items like cacao. Thus, the distance
from these fields to high population centres questions their role in broader Maya
subsistence.
The Yalahau region of the Yucatn Peninsula (Figure 5.1) provides another
area of study of wetland soil and Maya subsistence in its infancy. Like so many
other study areas, here the first clues come from surface alignments adjacent to
sites and wetlands. Fedick (2003) conducted a field survey of the El Edn wetland
area in Quintana Roo, Mexico, consisting in part of an intensive surface survey of
78 rock alignment features in this Yalahau region wetland. The survey team found
no prehistoric artefacts nor did they find Historic Period artefacts to date the rock
alignments, and they noted that dating such structures is necessarily problematic
(Fedick 2003, p.348). There was, however, a small Late Preclassic community
nearby that provided associated dates of 100bcad350/450 (Fedick 2003, p.349).
Pollen studies to identify and reconstruct native and cultivar plant history have
not yet been undertaken in the El Edn wetland. Fedick provided a description
for the wetlands soils: thin, sandy silts, and silty clay, about 20 cm over bedrock
[L]ower areas contain soils up to about 60cm deep with peaty deposits over
silty to sandy clay (2003, p.342). These coastal wetlands are of course prone to
any changes in groundwater flow and sea level change. The rock alignments of El
Edn appeared to not be functional as agricultural features under todays higher sea
and water table stands. They would only be functional under lower water levels,
which also helps us infer minimum ages for these features based on our patchy
knowledge of past sea levels. Fedick noted that the water table in this region was
likely 0.751m lower in the Late Preclassic than today (Fedick 2003). Because of
the difficulty of dating wetland features and reconstructing their hydrologic and
botanical histories clearly, physical evidence for bajo and terrace manipulation and
agriculture within the core of the Maya realm is far more extensive than for wetland
agriculture. Fedick and others suggest that the rock alignments might have partly
functioned to raise periphyton for fertilising (see Chapter 4).
103
Wetlands in the Americas
Bajo Agriculture
The difference between Bajo and wetland agriculture is that bajos karst depressions
are seasonally wet for four to six months in a year, whereas perennial wetlands
have water tables near the surface for most of the year. The soils of wetlands are
dominated by peats and marls (or histosols), and deposited or precipitated sediments,
whereas Bajos generally have vertisols or rendolls with thin patches of histosol (Beach
1998; Beach et al. 2003). Not all bajos fit into this classification; for example, the
large and well studied Bajo Morocoy in Quintana Roo does not, because it has
water table that is 1.5m from the surface in the dry season that makes it virtually
a perennial wetland through capillary water rise (Gliesmann et al 1983). This is a
bajo in which perennial wetland agriculture could have been practised.
Research since the 1930s, no doubt spurred on by their close proximity to
major sites, has hypothesised that bajos had been lakes or more extensive wetlands
during antiquity especially in the northern Petn of Guatemala (Figure 5.1). Indeed
in this region, Mirador, the largest Preclassic site, collapsed in the Early Classic
coincidental with the hypothesised drying of bajos. Some recent evidence suggests
these were more extensive wetlands and that parts, especially their margins, were
intensively used (Dunning et al. 2002; Hansen et al. 2002).
As noted above, it is significant that most of the major Maya sites had no or
few perennial wetlands nearby that could be likely wetland field sites: Tikal, Caracol,
Calakmul, Copn, Dos Pilas and Palenque; not to mention the northern Yucatn
sites like Chichen Itz, Chunchucmil and Uxmal (Figure 4.2). Hence, any descrip-
tion of the ancient Maya as an intensive wetland agricultural society is problematic,
and the other forms of agriculture must have been much more important. Several
current studies have therefore focused on the bajos as areas of agricultural and sil-
vicultural production. Studies have focused on excavation of field areas, diversions
and terraces, and on soil testing for proxy evidence of past crops (see Chapter 4).
In the central Petn around the Petexbatn, Seibal, and Cancuen, agriculture may
have been in small bajos on upland ridges. The heavily eroded backslope soils and
the thick build-up of sediments in the karst drains suggests heavy past use of these
areas (Dunning and Beach 2000).
One type of bajo agriculture occurs along the edges of bajos where many
footslope terraces and diversion chutes attempt to maintain soil and moisture
conditions. The footslope terraces may have had multiple purposes: catching sedi-
ment eroded from slopes, retaining sediments carted in from the wetlands, and
building up soil platforms above wet season inundation but within a favourably
moist microclimate and slope position (Beach et al. 2002 and 2003; Dunning et
104
Sheryl Luzzadder-Beach and Tim Beach
al. 2002; Hansen et al. 2002). Terrace studies show evidence of agriculture and soil
importation, possibly from bajos.
As has been done with wetland fields, studies are using more and more
proxies to study evidence of ancient agriculture in bajo soils. Maize and manioc
pollen or phytoliths at dateable soil layers indicates their propinquity to fields;
carbon isotopic ratios can indicate dominance by C4 plants like maize; and large
amounts of charcoal, phosphorus and artefacts probably indicate heavy human
use. Additionally, many other species show up in these pollen and other proxy
counts that are not food-related but nevertheless are still economic taxa such as
timber trees and thatch palms (Dunning et al. 2002; Dunning and Beach 2000;
Hansen et al. 2002).
The formation of some vertisols in bajos is another interesting aspect of soils
history. In many bajos, vertisols soils are thick and black throughout their verti-
cal profiles because self-ploughing of the expanding and contracting clays mixes
the entire profile. But in many other bajos, tiger vertisols develop vertical and
diagonal stripes of A-, B- and C-horizons, and in some others these vertical and
curved stripes are buried under typical horizontal soil layers (Beach et al. 2003).
There remain considerable questions about these patterns. In the first case, the
diagonal layers result from the long diagonal cracks that form in the dry season
when A-horizons, black topsoils, slough off into the cracks and fill them with
black soil in a matrix of grey C- and yellow-brown B-horizons. The descending
soil may also force lower horizons upward, thus further distorting vertisols into
melanges (Eswaran et al. 1999). In the bajos around La Milpa, Belize (Figure 4.2)
Vertisols have complicated natural and anthropogenic origins (Beach et al. 2003;
Dunning et al. 2002). Here vertisols that have 100cm deep diagonally oriented
horizons are topped by up to 70cm of typically horizontal horizons. These so-called
tiger vertisols originally formed as horizontal layers since they have a full suite of
horizons. Thus two major questions are, what causes them to change to diagonal
horizons, and why are the surface horizons horizontal? One explanation involves
the following historical sequences: (1) shallow rendoll soils formed by 3500bp;
(2) during the Maya Preclassic these soils aggraded from upland erosion by up to
100cm, and deforestation led to greater extremes in soil moisture and contraction
and expansion; (3) sediment loading on the clays provided a counterbalance mass
for swelling and shearing and ripping apart horizons; and finally (4) the upper
horizontal deposition occurred during and after the Classic Period, forming weakly
developed topsoils above the lower, deformed sequence. This later episode of lower
erosion and deposition corresponds to urbanisation, soil and water management
and abandonment at La Milpa.
105
Wetlands in the Americas
South America
In South America, most raised field features were apparently abandoned by the
time the Spanish arrived, because no written evidence was presented by the Spanish
chroniclers of the time (and only a few observations later) to suggest these raised
fields were under cultivation (Denevan 1970, 2002; Parsons 1989, p.205). There
were likely a very small number of raised fields still being used in the late nine-
teenth century (Erickson 2000), though most of these agroecosystems in the Rio
Catari basin near Lake Titicaca (Figure 5.1) seem to have been abandoned prior
to ad1150 (Kolata 2000, p.172).
Andean Highlands
The broader group of raised fields around the margins of Lake Titicaca and those
of the Andean Highlands of Peru provide South Americas best known examples
of wetland agriculture (Figure 5.1) (Denevan 1970, 2002; Erickson 2000; Kolata
2000). Raised fields occupy these waterlogged edges in regional conjunction with
agricultural terraces and a variety of other earth works, house mounds and causeways
(Parsons and Denevan 1989; Parsons and Bowen 1966; Erickson 2000). Human
influence in this landscape far precedes this, to at least 8000bp (Erickson 2000),
but specific dates for wetland fields are from about 1800bc and later. In a study of
the Lake Wiyamarka arm of Lake Titicaca, (the Rio Catari basin), Kolata dated
106
Sheryl Luzzadder-Beach and Tim Beach
the building of most raised fields from ad 600 to 1100 (Kolata 2000, p.172).
Erickson (2000, p.336) added that it was likely that the earliest farming on raised
fields was between 1800 and 900bc, and that it was developing along the lake
edges between 900bc and ad200, and was found more extensively in the area
between 200bc and ad600.
The raised fields of the Lake Titicaca Basin consist of alternating soil
platforms and canals. The platforms are variable in size, and range in shape from
strips to island-like structure surrounded by ditches (Erickson 2000). Their pat-
terns include several kinds of raised field geometries: open checkerboard, irregular
embanked, riverine, linear, ladder and combed field patterns (Denevan 2002,
pp.25862; Parsons and Denevan 1989, p.218). These field types all represent
massive transformations of this landscape, since their sizes range from 4 to 10m
wide, 10 to 100m long, and 0.5 to 3m tall (Erickson 2002, p.333). The soils of
these structures also show manipulation to a depth of 2m or more on a massive
regional scale (Erickson 2000, p.347). Excluding all other human earth works in
this region, and there are many, these agricultural raised fields alone cover some
120,000ha, an area about one-half the size of Belgium (Erickson 2000).
Constructed lowland wetland field structures of the Titicaca Basin can be
classified in two forms: raised fields and sunken gardens, or qochas (Erickson 2000).
Qochas coexist with raised fields, and are used even today (Erickson 2000, p.337).
Qochas are shallow depressions, ranging in size from 0.1 to 4hectares. They are
found in poorly drained areas, and are thought to be largely natural formations,
but are linked by artificial canal networks (Erickson 2000, p.338). These wetlands
are still used today for a variety of crops and animal forage and they are also docu-
mented in early colonial writings, but their ancient and historical uses are uncertain
(Erickson 2000, p.341).
Given the tremendous labour investments implied in these earth works, they
must have been an integral part of the food supply for the region (Denevan 2002).
In reviewing population estimates made in various studies from 1982 to present,
Denevan (2002) estimated that Titicacas raised fields supported between 215,000
and 2.29 million people. Estimating populations is notoriously error-prone but
even the least of this order of magnitude range is substantial.
Excavating, building and periodically maintaining these fields and ditches
has left a complicated and lasting human imprint on soils development, structure
and stratigraphy. Wetland fields may provide the quintessential palimpsest because
they may have been worked and abandoned at several different times.
Few published soils studies existed when Smith, Denevan and Hamilton
(19661968 regional survey) explored the raised fields of Titicaca (Figure 5.1) (Smith
et al. 1968; Denevan 2002). One study covered an area north of Puno, in the Pau-
carcolla-Juliaca plain, and classified the soils as gley, humic Andean Planosols (see
United Nations [FAO/UNESCO] 1974). Denevan reported the soils to be poorly
107
Wetlands in the Americas
drained lacustrine (lake deposit) soils of fine sediment size with moderate fertility,
though in some places strongly alkaline (Denevan 2002, pp.2623). Raised field
construction attempted to overcome the two main limitations of these soils: high
alkalinity and a high water table. Since alkalinity decreased down these soil profiles,
building drained fields essentially turns over the soil, exposing soils of lower pH
and potentially higher fertility on the raised planting beds. Building up the fields
and canalising drainage was also an effort to overcome poor drainage, obviously
the major limitation for agriculture in swamps (Denevan 2002).
during the last 3,000 years (van der Hammen 1974, pp.125; van der Hammen
and Hooghiemstra 2003).
North America
2000). Doolittle noted that this highly productive flood recessional agriculture
dominated the lower courses of all the rivers from the Yaqui to the Colorado in
the south-west, and Spanish settlers quickly recognised their value and colonised
these productive lands (Doolittle 2000, p.427).
(1969, p.365), and he hypothesised that this region was influenced by expansion
northward of Mesoamerican civilisation frontiers (1969, p.365). But there
remains no hard evidence linking Mesoamerican culture and its various forms of
intensive agriculture with Cahokia. The evidence for raised field agriculture itself at
Cahokia is still considered sparse, compared with other North American sites such as
the Parker Earthwork in Ontario (Doolittle 2000). Since Cahokia was so large, the
spotty evidence for intensive raised fields here parallels the situation for the Maya
Lowlands, where wetland fields are also distant from the main population centres.
The possible functions of the raised or ridged field sites in the Mississippian culture
might have included making weed removal easier, delivering irrigation, reclaiming
bottom land, draining cold air, and raising plants above water-logged soils (Fowler
1969, p.374; Doolittle 2000, p.441). But some of the ridged fields were located
higher than the saturated flood plain zones (Fowler 1969, p.374). Regardless of
site, the fields represent an efficient strategy for increasing food production for the
Mississippian temple-town type of cultural expansion (Fowler 1969, p.374).
Gartner (1999) has offered further insight into the function and meaning
of Pre-Columbian raised fields in Wisconsin. As noted above, some but not all of
the raised fields in the eastern North American landscapes were found in valley
bottom and wetland contexts. The raised/ridged fields perhaps had a socio-political
function in addition to a practical function (Gartner 1999), which may explain
their more wide-ranging geomorphic settings. For example, the Hulbert Creek fields
site in Wisconsin was located in a less than ideal choice for agriculture because of
thinner soils (Gartner 1999, p.675). These raised or ridged fields are highlighted
in a study of the Oneota people, who practised agriculture near Lake Koshkonong
between ad1000 and ad1150 (Gartner 1999, p.672). They functioned physically
in ways similar to raised fields elsewhere in the Americas, offering frost protection,
improved soil fertility, disease protection, well-aerated and well-drained soil, soil
moisture storage, minimised risk, and maximised solar exposure (by orienting the
planting beds in a north-south direction creating a warmer soil) (Gartner 1999,
pp.674, 682). But these ridged fields may also have powerfully symbolic meaning.
The raised fields can be viewed, along with effigy mounds and other earth works,
as landscape elements of territoriality (Gartner 1999, p.672). Since ridged fields
came into production as mound building ceased, they may constitute the symbolic
successor for Mounds in Oneota societies, and represent communal labour and
a commitment to place (Gartner 1999, p.681). Ridged field motifs even made
their way onto Pre-Columbian Oneota pottery of the time. Some motifs were very
similar to plan view projections [and] topological sequence[s] of the adjacent
ridged fields (Gartner 1999, p.681).
111
Wetlands in the Americas
Summary
There can be no doubt that many wetland soils of the Americas have considerable
anthropogenic inputs, and the evidence for human and environmental transforma-
tion of wetlands over the last millennia is overwhelming. In turn, human culture
in this region has not developed independently from its environment, but in a
coupled, complex system with the environment (Kolata 2000). The best evidence
for intensive wetland soil manipulation comes from the Basin of Mexico and Lake
Titicaca in South America. Here we have ethnographic, historical and archaeologi-
cal evidence for intensive wetland agricultural systems and the anthropogenic soils
within them.
The stories of North America and the Maya Lowlands are based more on
archaeological and geomorphologic excavations, and are thus more difficult to
ascertain and interpret. North America provides a surprisingly complex picture
of human participation in the subsistence landscape in its Pre-Columbian ridged
field and flood recessional cultivation systems. Here, wetland soil manipulation
occurred on a smaller scale, though a high proportion of its wetlands have been
altered and their evidence compromised.
For the Maya Lowlands, wetland manipulation for agriculture certainly oc-
curred, but the evidence for intensive agriculture near the great population centres
is lacking. Here, the last thirty years of Maya soil research has converged around
a consensus that the Maya relied on a mosaic of soil-agroecological environments
for subsistence. They exploited wetlands, but more so bajos, sinkholes, terraces,
floodplains and uplands with varying intensity. The Maya soil environment also
experienced much natural evolution from sea level, groundwater level, and droughts
and other climate changes. The Maya soil environment was also altered by human-
induced erosion and sedimentation, and possibly deforestation-induced increased
runoff and water table rise. One verified impact on wetland soils in the Maya
Lowlands was recurrent drought, especially in the Classic period. This is a major
finding on the Maya environment and possibly on Maya civilisation, but we still
need to model how these droughts, especially the Late Classic drought, affected
wetlands, soils and subsistence (Hodell et al. 1995, 2000 and 2001; Haug et al.
2003). Another unverified but potential impact on bajos, floodplains, and wetlands
in the Maya Lowlands was the impact of deforestation on runoff and water tables
(Ray et al. 2005).
In all regions we need more focused fieldwork that uses all possible forms
of evidence to understand the history and geography of these human manipulated
wetland systems. As borne out in Maya lowlands wetland research, we especially need
to appreciate the role of water chemistry in wetlands, and to consider all relevant
environmental and human input factors. Patterned fields have been brought to our
112
Sheryl Luzzadder-Beach and Tim Beach
attention largely and serendipitously through early aerial photography. These fields
promise to reveal much more about ancient subsistence, changing environments
and potential development through further excavation, soil studies, restoration
efforts and the many forms of proxy evidence.
But provisionally it is fair to conclude that over several millennia the peoples
of the Americas, in hundreds of distinct locations, used their ingenuity and labour
to alter soils and hydrology in their quests for food and fibre. Around and within
a range of wetland environments, they ditched, drained, diverted and raised fields.
Like agricultural peoples everywhere, they made their livings from soil and water,
and as a result never left them just as they found them.
References
Armillas, Pedro. 1971. Gardens on Swamps. Science, 174: 65361.
Beach, Timothy. 1998. Soil Catenas, Tropical Deforestation, and Ancient and Contemporary
Soil Erosion in the Petn, Guatemala. Physical Geography, 19(5): 378405.
Beach, Timothy, and Sheryl Luzzadder-Beach. 2002. Pre-Columbian Aggradation and
Wetland Agriculture in Belize. Paper presented at 2002 AAG Annual Meetings, Los
Angeles CA, March 2002.
Beach, Timothy, Sheryl Luzzadder-Beach, Nicholas Dunning, Jon Hageman and Jon Lohse.
2002. Upland Agriculture in the Maya Lowlands: Ancient Conservation in Northwestern
Belize. The Geographical Review, 92(3): 37297.
Beach, Timothy, Nicholas Dunning, Sheryl Luzzadder-Beach and Vernon Scarborough.
2003. Depression Soils in the Lowland Tropics of Northwestern Belize: Anthropogenic
and Natural Origins. In A. Gomez-Pompa, M. Allen, S. Fedick and J.J. Jimenez-Os-
ornio (eds), The Lowland Maya Area: Three Millennia at the Human-Wildland Interface
(Binghamton, NY: Haworth Press), pp.13974.
Beach, T., N. Dunning, S. Luzzadder-Beach, J. Lohse and D. Cook. 2006. Ancient Maya
Impacts on Soils and Soil Erosion in the Central Maya Lowlands. Catena, in press.
Bloom, Paul R., Mary Pohl, Cynthia Buttleman, Frederick Wiseman, Alan Covich, Charles
Micksicek, Joseph Ball and Julie Stein. 1983. Prehistoric Maya Wetland Agriculture
and the Alluvial Soils near San Antonio Rio Hondo, Belize. Nature, 301: 41719, 3
February 1983.
Carrasco, David and Scott Sessions. 1998. Daily Life of the Aztecs: People of the Sun and
Earth. Westport, CT: Greenwood Press.
Chapin, Mac. 1988. The Seduction of Models: Chinampa Agriculture in Mexico. Grassroots
Development, 12(l): 817.
Coe, Michael D. 1964. The Chinampas of Mexico. Scientific American, 211(1): 908,
July issue.
Crossley, Phil L. 2004. Just Beyond the Eye: Floating Gardens in Aztec Mexico. Historical
Geography, 32: 11135.
113
Wetlands in the Americas
Dalan, Rinita A., George R. Holley, William I. Woods, Harold W. Watters, Jr. and John
A. Koepke. 2003. Envisioning Cahokia: A Landscape Perspective. Dekalb, IL: Northern
Illinois University Press.
Denevan, William M. 1970. Aboriginal Drained-Field Cultivation in the Americas. Science,
169(3946): 64754, 14 August 1970.
Denevan, William M. 1992. The Pristine Myth: The Landscape of the Americas in 1492.
Annals of the Association of American Geographers, 82(3): 36985.
Denevan, William. 2002. Cultivated Landscapes of Native Amazonia and the Andes. Oxford:
Oxford University Press.
Doolittle, William E. 1992. Agriculture in North America on the Eve of Contact: A Reas-
sessment. Annals of the Association of American Geographers, 82(3): 386401.
Doolittle, William E. 2000. Cultivated Landscapes of North America, Oxford Geographical
and Environmental Studies. Oxford: Oxford University Press.
Dunning, N., and T. Beach. 2000. Stability and Instability in Pre-Hispanic Maya Landscapes.
In D. Lentz (ed.), Imperfect Balance: Landscape Transformations in the Precolumbian
Americas (New York: Columbia University Press), pp.179202.
Dunning, N., and T. Beach. 2004. Noxious or Nurturing Nature? Maya Civilization in En-
vironmental Context. In C. Golden and G. Borgstede(eds), Continuities and Changes in
Maya Archaeology: Perspectives at the Millennium (New York: Routledge), pp.12541.
Dunning, Nicholas P., Sheryl Luzzadder-Beach, Timothy Beach, John G. Jones, Vernon
Scarborough and T. Patrick Culbert. 2002. Arising from the Bajos: The Evolution of a
Neotropical Landscape and the Rise of Maya Civilization. Annals of the Association of
American Geographers, 92(2): 26783.
Erickson, Clark L. 1992. Prehistoric Landscape Management in the Andean Highlands:
Raised Field Agriculture and its Environmental Impact. Population and Environment,
13: 285300.
Erickson, Clark. 1998. Applied Archaeology to Rural Development: Archaeologys Poten-
tial Contribution to the Future. In Michael B. Whiteford and Scott Whiteford (eds),
Crossing Currents: Continuity and Change in Latin America (Upper Saddle River, NJ:
Prentice Hall), pp.3445.
Erickson, Clark L. 2000. The Lake Titicaca Basin: A Precolumbian Built Landscape. In David
L. Lentz (ed.), Imperfect Balance: Landscape Transformations in the Precolumbian Americas,
The Historical Ecology Series (New York: Columbia University Press), pp.31156.
Eswaran, H., F. Beinroth, P. Reich and L. Quandt. 1999. Vertisols: Their Properties,
Classification, Distribution, and Management. In Monograph of World Soil Resources
(Washington, D.C: Natural Resources Conservation Service, USDA).
Fedick, Scott L. 2003. Archaeological Evidence for Ancient and Historic Resource Use
Associated with the El Edn Wetland, Northern Quintana Roo, Mexico. In A. Gomez-
Pompa, M. Allen, S. Fedick and J. J. Jimenez-Osornio (eds), The Lowland Maya Area:
Three Millennia at the Human-Wildland Interface (Binghamton, NY: Haworth Press),
pp.33959.
Fowler, Melvin L. 1969. Middle Mississippian Agricultural Fields. American Antiquity,
34(4): 36975.
114
Sheryl Luzzadder-Beach and Tim Beach
Gartner, William Gustav. 1999. Late Woodland Landscapes of Wisconsin: Ridged Fields,
Effigy Mounds and Territoriality. Antiquity, 73: 67183.
Gliessman, S.R. 1991. Ecological Basis of Traditional Management of Wetlands in Tropi-
cal Mexico: Learning from Agroecosystem Models. In M.L. Oldfield and J.B. Alcorn
(eds), Biodiversity: Culture, Conservation and Ecodevelopment (San Francisco: Eastview
Press).
Gliessman, Stephen R., B.L. Turner II, F.J. Rosado May and M.F. Amador. 1983. Ancient
Raised Field Agricuture in the Maya Lowlands of Southeastern Mexico. In J.P. Darch
(ed.), Drained Field Agriculture in Central and South America, BAR International Series
189 (Oxford UK: British Archaeologial Series).
Gomez-Pompa, Arturo, Hector Luis Morales, Epifanio Jimenez Avilla and Julio Jimenez
Avilla. 1982. Experiences in Traditional Hydraulic Agriculture. In Kent V. Flannery
(ed.), Maya Subsistence (New York: Academic Press), pp.32742.
Hansen, Richard D., Steven Bozarth, John Jacob, David Wahl and Thomas Schreiner. 2002.
Climatic and Environmental Variability in the Rise of Maya Civilization: A Preliminary
Perspective from Northern Peten. Ancient Mesoamerica, 13(2): 27395.
Harrison, Peter D. 1996. Settlement and Land Use in the Pulltrouser Swamp Archaeological
Zone, Northern Belize. In Scott L. Fedick (ed.), The Managed Mosaic: Ancient Maya
Agriculture and Resource Use (Salt Lake City: University of Utah Press), pp.17790.
Haug, G.H., D. Gnther, L.C. Peterson, D.M. Sigman, K.A. Hughen and B. Aeschliman.
2003. Climate and the Collapse of Maya Civilization. Science, 299: 17315.
Hebda, Richard. J., Alfred H. Siemens and Alistair Robertson. 1991. Stratigraphy, Depositional
Environment, and Cultural Significance of Holocene Sediments in Patterned Wetlands
of Central Veracruz, Mexico. Geoarchaeology: An International Journal, 6(1): 6181.
High, L.R. 1975. Geomorphology and Sedimentology of Holocene Coastal Deposits,
Belize. In K.W. Wantland and W.C. Pussey III (eds), Belize Shelf-Carbonate Sediments,
Clastic Sediments, and Ecology (Tulsa, OK: American Association of Petroleum Geolo-
gists) pp.5396.
Hodell, D.A, M. Brenner and J.H. Curtis. 1995. Possible Role of Climate in the Collapse
of Classic Maya Civilization. Nature, 375: 3914.
Hodell, D.A., M. Brenner and J.H. Curtis. 2000. Climate Change in the North American
Tropics and Subtropics since the Last Ice Age: Implications for Environment and Culture.
In D.L. Lentz (ed.), An Imperfect Balance: Landscape Transformations in the Precolumbian
Americas (New York: Columbia University Press), pp.1318.
Hodell, D., M. Brenner, J. Curtis and T. Guilderson. 2001. Solar Forcing of Drought
Frequency in the Maya Lowlands. Science, 292: 136770.
Jacob, John S. 1995. Ancient Maya Wetland Agricultural Fields in Cobweb Swamp, Belize:
Construction, Chronology, and Function. Journal of Field Archaeology, 22: 17590.
Jacob, J.S., and C.T. Hallmark. 1996. Holocene Stratigraphy of Cobweb Swamp, a Maya
Wetland in Northern Belize. Geological Society of America Bulletin, 108(7): 88391.
Kolata, Alan L. 2000. Environmental Thresholds and the Natural History of an Andean
Civilization. In Garth Bawden and Richard Martin Reycraft (eds), Environmental Disaster
115
Wetlands in the Americas
Pope, Kevin O., Mary D. Pohl and John S. Jacob. 1996. Formation of Ancient Maya
Wetland Fields: Natural and Anthropogenic Processes. In Scott L. Fedick (ed.), The
Managed Mosaic: Ancient Maya Agriculture and Resource Use (Salt Lake City: University
of Utah Press), pp.16576.
Purdy, Barbara A. (ed.). 2001. Enduring Records: The Environmental and Cultural Heritage
of Wetlands. Oxford: Oxbow Books.
Ray, D.K., U.S. Nair, R.M. Welch, R.O. Lawton, R.J. Oglesby, R.A. Pielke, T.A. Sever and
D. Irwin. 2005. The Role of Deforestation in the Collapse of Classic Maya Civiliza-
tion: Lessons for the Current Land Use Management in Northern Mesoamerica. EOS
Transactions AGU 86(52), Fall Meeting Supplement, Abstract B33F-07.
Ricketson, Oliver Jr. and A.V. Kidder. 1930. An Archaeological Reconnaissance by Air in
Central America. The Geographical Review, 20(2): 177206.
Schmidt, Peter J. 1977. Un Sistema de Cultivo Intensivo en la Cuenca del Rio Nautla, Ve-
racruz. Boletin del Instituto Nacional de Anthropologia e Historia, Epoca 3(20): 5060.
Siemens, Alfred H. 1978. Karst and the Pre-Hispanic Maya in the Southern Lowlands. In
P. D. Harrison and B.L. Turner II (eds), Pre-Hispanic Maya Agriculture (Albuquerque
NM: University of New Mexico Press), pp.11743.
Siemens, Alfred H. 1982. Pre-Hispanic Use of the Wetlands in the Tropical Lowlands of
Mesoamerica. In K. Flannery (ed.), Maya Subsistence: Studies in Memory of Dennis E.
Puleston (NY: Academic Press), pp.20525.
Siemens, Alfred H. 1983. Wetland Agriculture in Pre-Hispanic America. Geographical
Review, 73(2): 16681.
Siemens, Alfred H. 1998. A Favored Place. Austin TX: University of Texas Press.
Siemens, A.H., and D.E. Puleston. 1972. Ridged Fields and Associated Features in
Southern Campeche: New Perspectives on the Lowland Maya. American Antiquity,
37: 22839.
Siemens, Alfred H., Richard J. Hebda, Mario Navarette Hernandez, Dolores R. Piperno,
Julie Stein and Manuel G. Zola Baez. 1988. Evidence for a Cultivar and a Chronology
from Patterned Wetlands in Central Veracruz, Mexico. Science, 242(4875): 1057.
Sluyter, Andrew. 1994. Intensive Wetland Agriculture in Mesoamerica: Space, Time and
Form. Annals of the Association of American Geographers, 84(4): 55784.
Smith, Clifford T., William M. Denevan, and Patrick Hamilton. 1968. Ancient Ridged
Fields in the Region of Lake Titicaca. The Geographical Journal, 134: 35367.
Smith, Michael E. 2003. The Aztecs, 2nd edn. Malden MA: Blackwell Publishing.
Soil Survey Staff. 1998. Keys to Soil Taxonomy, 8th edn. Washington DC: USDA Natural
Resource Conservation Service/U.S. Government Printing Office.
Steggerda, M. 1941. Maya Indians of Yucatn, Publication No. 531. Washington DC:
Carnegie Institution of Washington.
Terry, Richard E. 2003. Stable Carbon Isotopes of Soil Organic Matter Hold a Record
of Ancient Maize Agriculture in Mesoamerica (Presentation). In Giovanni Boschian
(ed.), Second International Conference on Soils and Archaeology Pisa, 12th-15th May,
2003, Extended Abstracts (Pisa Italy: University of Pisa and International Union of Soil
Sciences), pp.1256.
117
Wetlands in the Americas
Turner, B.L. II. 1974. Prehistoric Intensive Agriculture in the Mayan Lowlands. Science,
185(4146): 11824, 12 July 1974.
Turner, B.L. II. 1983. Once Beneath the Forest: Prehistoric Terracing in the Rio Bec Region of
the Maya Lowlands. Boulder, CO: Westview.
Turner, B.L. II and Peter D. Harrison, 1981. Prehistoric Raised-field Agriculture in the
Maya Lowlands. Science, 213(4506): 399405.
Turner, B.L. II and Peter D. Harrison (eds). 1983. Pulltrouser Swamp: Ancient Maya Habitat,
Agriculture, and Settlement in Northern Belize. Austin TX: University of Texas Press.
United Nations (FAO/UNESCO). 1974. FAO/UNESCO Soil Map of the World: Legend.
Paris, France: UNESCO. Online at: http: //www.fao.org/ag/agl/key2soil.stm , FAO,
2004.
van der Hammen, Thomas. 1974. The Pleistocene Changes of Vegetation and Climate in
Tropical South America. Journal of Biogeography, 1(1): 326, March 1974.
van der Hammen, Thomas and Henry Hooghiemstra. 2003. Interglacial-Glacial Fuquene-3
Pollen Record from Colombia: an Eemian to Holocene Climate Record. Global and
Planetary Change, 36(2003): 18199.
West, Robert C. and John P. Augelli. 1966. Middle America: Its Land and Peoples. Englewood
Cliffs, NJ: Prentice-Hall.
Whitmore, T. and B. Turner II. 2001. Cultivated Landscapes in Mesoamerica on the Eve of
the Conquest. Oxford: Oxford University Press.
Wilk, R.R. 1981. Agriculture, Ecology, and Domestic Organization among the Kekchi
Maya. PhD Dissertation, University of Arizona.
Woods, William I. (ed.). 1992. Late Prehistoric Agriculture: Observations from the Midwest,
Studies in Illinois Archaeology No. 8. Springfield IL: Illinois Historic Preservation
Agency.
118
Kate B. Showers
Kate B. Showers
Introduction
African soils cannot be separated from African history because they are, in themselves,
historical bodies and they have interacted with human history since its beginning.
The history of soil other than of its genesis (a concern of pedology) is a history of
its perceptions and the consequences of its use. Such a history must necessarily be
one of exploration and discovery, of experimentation and classification. The late
nineteenth- and twentieth-century European experience is only the latest in a long
series of human encounters with the continent. An essential feature of Africas soils
is their diversity. This is to be expected, when one considers the sheer size of the
continent, and its many ecosystems. Africas approximately 30,221,392 sq.km. is
greater than the land areas of China, Europe and the United States combined, see
Figure 6.1. It accounts for 20.4 per cent of the earths land (National Geographic
Society 1990, p. 130). This vast area contains snow-capped mountains, hot deserts,
rivers and wetlands with diverse vegetation as shown in Figure 6.2.
The diversity of African soils was demonstrated in the twentieth century
by the many complications encountered when modern or improved agricultural
techniques were implemented throughout the continent. African soils simply did
not fit preconceived notions, and did not respond in a manner similar to soils in
other places (Lal and Sanchez 1992). Purveyors of modern and improved agricul-
ture were, to use T. Beachs (1998, p. 400) concept, environmental pioneers, as
had been Africans before them when they migrated and settled in different parts
of the continent.
Europeans first encountering the tropical African high forest concluded that
the red-coloured soils must be extremely fertile to support such luxuriant growth.
In time, red soils became known as the curse of Africa, liable to harden irreversibly
and become useless under cultivation. Eventually, all colours of soil were said to
119
A History of African Soil
Figure 6.1. The land mass of Africa exceeds that of Europe, the USA and China combined.
Based on How big is Africa?, Trustees of Boston University n.d..
erode with ferocity, when not blowing in desiccating winds. Finally, it was said that
African soils were worn out by African smallholders agricultural mining of their
fertility, and destroyed by trampling of their livestock. To gain some perspective,
this chapter will attempt to view African soils as understood and used by Africans
and Europeans for as much of the past as there is some kind of evidence.
Charting the ideas and activities of various waves of environmental pioneers
on the African continent is a complicated process. Africa remains an understudied
continent, and African knowledge systems a profoundly endangered resource
120
Kate B. Showers
Tropical rainforest
(broadleaf evergreen)
East African coastal forest
Deciduous forest-woodland
savanna
Brush-grass savanna
Steppe (grass, brush and
thicket)
Semi-desert
Desert
Mediterranean evergreen
forest- hard-leaf scrub
Montane forest-tundra
1000km
are generally unknown to the larger world community. The historical record (in
its broadest sense) is incomplete. Few documents have been analysed from an earth
perspective, oral histories of societies and their landscapes have not been collected
in most locations, and archaeological research has been very limited. However,
materials from a wide range of sources do exist from which examples can be drawn
for consideration. While the absence of reliable evidence is problematic, the confu-
sion created by apparently authoritative documentation that lacks a firm factual
basis must also be clarified. The twentieth century was a time when overlapping
knowledge systems and practices existed, but not necessarily to the benefit of the
121
A History of African Soil
newly arrived environmental pioneers or the earth. Prejudice and ignorance often
prevented information exchanges.
To provide a base for considering a history of African soil, general discussions
of the perceptions of Africa and its soils, measurable properties of African soils, and
soil knowledge systems will be presented. Attention will then turn to examples of
African and European encounters with African soils. Four categories of activities
that could have affected soil properties or entire soil bodies will be considered: crop
production, livestock management, architecture/urbanisation, and mining. Discus-
sion will centre around specific studies of particular places or systems in an attempt
to identify mechanisms of change, rather than a continuous continent wide analysis
in careful chronology. Generalisations from relevant studies will be made hesitantly,
keeping in mind Richards (1985, pp. 12, 13) cautions about working from a base
of ecological particularism and seeking to avoid both glib generalisation and ad hoc
arguments or special pleading, as well as concerns about the distortions inherent
in extrapolation. Topics in need of further consideration or research will be noted,
as this chapter is intended as a preliminary survey of the topic.
1
Discussions not based on crisis narratives are often accused of asserting the idea of Merrie
Africa. In defence against such a charge, Helge Kjekshus reminded readers in a footnote
to the Introduction to the 1996 impression of his 1976 classic work, Ecology Control and
Economic Development in East African History, The term Merrie Africa is first used by
Hopkins (1973: 10) as a contrast to Primitive Africa of Alfred Marshall (1938) to set
the stage for Hopkins pioneering work on West African economic history. As the reader
will recall, Merrie Africa denoted to Hopkins a mythical situation of populations living,
without working, in abundance and plenty, but pursuing lives of ease and leisure of which
major parts consisted of interminable dancing and drumming. Kjekshus then states that
in his book, The portrait of East Africans that emerges ... is that of hard workers, skillful
planners, active learners and ultimate survivors. These are clearly not characteristics of the
cast that made up A.G. Hopkins myth of Merrie Africa (Kjekshus 1996, p.xxix).
123
A History of African Soil
narratives, sources of information must be identified that can suggest the nature
of interactions between humans and soils, as well as indicate whether they resulted
in transformation, creation or destruction of soil bodies. A not inconsequential
problem, as Scoones (1997) points out and Chambers (1983) before him is
that even reliable studies are simply snapshots of a particular place at a specific
time. In Africa, some people in some places at some times experienced events and
conditions that didnt happen elsewhere, or at that place, but in a different time.
Much of Africa has an extremely variable rainfall distribution, making it difficult for
anyone to identify what is typical and what is extreme. In many cases, the extremes
(drought and flood) are both typical and normal.
But it is the very perception of normal that makes historical documents
difficult. As discussed by Showers (1989), normal to most people is that which is
familiar, not a statistical concept of mean and deviation. Most documenters of Africa
travellers, residents, or experts found whatever part of Africa they encountered
to be quite different from their place of origin, and judged it to be abnormal. West
Africas climate and the presence of malaria inhibited European settlements, but
not commerce or the establishment of plantations. The ubiquitous red soils were
very distinct, even to short-term visitors. In some places, red soils cleared for crops
hardened. These soils were described as laterite, the name coined by British geolo-
gist F. Buchanan-Hamilton in 1807 for red earth (it was not referred to as soil) that
hardened and was used for bricks in the Kerala District of India (Russell 1973, p.
730; Schantz and Marbut 1923, pp. 118, 18082). In 1820 P. Berthier analysed
soil samples from the Fouta Djallon of West Africa, and found them to be similar
to that of Kerala, and in 1911 J.D. Falconer described some Nigerian soils as iron
clay (Encyclopaedia Britannica 1929 vol. 13, p. 740). Soon all red West African
soils were assumed to be laterite.
Semi-arid or arid regions of Africa were frequently perceived as being defec-
tive, or in a state of decline or degradation. These conditions were often blamed
on local people and their land use system. In the Cape Colony (modern South
Africa), R. Grove (1989) first argued, and Endfield and Nash (2002) documented
in detail, how fundamentalist interpretations of the Bible led missionaries to believe
that the dry landscape was a sign of Gods punishment of the indigenous people;
in neighbouring Basutoland (modern Lesotho), the relative lack of trees in the
natural grassland was assumed by some to have been due to deforestation by the
local residents (Showers 1989; Showers 2005); and in French North Africa, par-
ticularly Morocco, the lack of trees was said to be proof of the destructiveness of
Arab land-use practices (Davis 2005). In both the semi-arid West African Sahel and
southern Africa there is a literature attributing the low fertility status of the soils to
traditional land use practices, particularly the burning of grass or dung.
Ideology, bias and perceptions of normal clearly have distorted observers
accounts of Africa. But these predispositions may well have been magnified by a
124
Kate B. Showers
2
When considering environmental factors in an historical context, historians are confronted
by two problems: the belief of some post-modernists that science is just a social construc-
tion, and a charge of environmental determinism. Those concerned with these ideas might
consider Nancy Jacobs discussion of the distinction between nature, a social construction,
and an actually existing nonhuman biophysical world with its own integrity (Jacobs 2003,
p.19). Jan Vansina (1990) argues that historians underestimation of this diversity [within
a rainforest] leads to a flawed understanding of the interactions between people and their
habitats (Vansina 1990, p.41). Consideration of the biophysical world (of which soils are a
fundamental part) is, therefore, not environmental determinism but, rather, an opportunity
to look at the ways in which people engage with their surroundings, to see previously ignored
expressions of creativity, and to more fully appreciate local environmental knowledge.
125
A History of African Soil
resulting in soil types unknown in the northern hemisphere (Eswaran et al. 1992,
p. 3). Since the African continent lies in the tropics and subtropics, except in high
mountain areas, the activities of flora and fauna are uninterrupted by cold, so that
soil forming processes can occur year round. Soils in Africa more resemble those
of Brazil, peninsular India and Australia than those of Europe or North America
(Grove, A.T. 1970, p. 26).
The only statement that can be made with certainty about African soils is
that they are enormously diverse. For every general statement made about African
soil conditions, an example to the contrary can be found. With this in mind, let
us consider some broad characterisations. According to the documentation ac-
companying the Africa sheet of the FAO/Unesco Soil Map of the World (1977),
many African soils are highly weathered and have lower plant nutrient contents
than northern hemisphere soils. Levels of phosphorus are low enough in much of
the continent to limit plant growth. Elevated levels of aluminium and manganese,
which can inhibit root growth of many plants, are also common in highly weathered
soils. In some places the iron content approaches that of low-grade ore. Sub-Saharan
West African soils have their fertility rejuvenated each year when dust, laden with
basic cations (calcium, magnesium, potassium and sodium), blows in with the
Harmattan winds. Arid and semi-arid regions of the north, east and south of the
continent have sandy, stony and/or rocky soils that can be quite shallow (FAO/
Unesco 1977, p. viii). Surface crusting and sealing, as well as salt deposition, are
common in these soils. Areas of moving sand dunes exist in both the Sahara and
Namib Deserts. Layers of dense or hard soil, called pans, can be found below the
surface of soils in all rainfall regimes.
Many of Africas highly weathered clay soils have a red colour, signifying
high iron content (Eswaran 1988, p. 6). The fate of this iron depends upon soil
conditions. It may simply accumulate as a stain on the surface of soil particles, or
it can be concentrated. Where the water table fluctuates either locally or on the
scale of a floodplain the soil solution can become enriched with iron, which is
subsequently deposited as mottles, concretions, nodules or continuous sheets 25
metres or even 10 metres thick (Eswaran 1988, p. 7; Buckle 1978, p. 68). Re-
ferred to variously as laterite or plinthite, this material is characteristically soft and
permeable when moist, but hardens upon drying. (Note: the word laterite has been
used so broadly that its precise meaning is unclear. For this reason, some pedolo-
gists and soil scientists have refused to use the term. However, forms of the word
are components of some international soil classification systems [see Faniran and
Areola 1978, p. 161; Eswaran et al. 1992, p. 7]). The extent of hardness achieved,
and its irreversibility, varies (Russell 1973, p. 731). When hard, it can be referred
to as ferricrete, lateritic duricrust, or petroplinthite (Buckle 1978, p. 68; Eswaran
1988, p. 7). Soils with lateritic or plinthic layers have developed under both forest
and savanna in West Africa, and are extensive in regions with annual rainfalls of
126
Kate B. Showers
genetic soil science. The idea of factors of soil formation was later expanded upon
in the United States of America and expressed in the simplified flower diagram
shown in Figure 6.3.
Dokuchaevs ideas reached those who could only read English when The
Great Soil Groups of the World and Their Development, written by his student, D.K.
Glinka, was translated from German to English by C.F. Marbut and published in
1927 (Lipman 1933; Soil Survey Staff 1951, p. 3; Faniran and Areola 1978, p. 6).
A variety of national soil classification systems were subsequently devised based
upon physical and chemical properties of soils as revealed by soil profile analysis.
Characteristics such as colour, texture (amounts of sand, silt and clay), pH (a
measure of acidity), base status (amounts of calcium, magnesium, sodium and
potassium), water relations, and position in the landscape were used as determin-
ing characteristics.
129
A History of African Soil
Soil classification and mapping has been an interest in Europe and North
America since the turn of the twentieth century. According to the FAO/Unesco
Soil Map of the World, two significant early soil maps of the African continent
were those of C.F. Marbut in 1923 and Z.Y. Shokalskaya in 1944. Both were
general maps of hypothesised soil distribution based on climatic, lithological or
phytogeographical factors. Working in East Africa, chemist G. Milne used local soil
names and knowledge in his early studies of soil catenas (sequences of soils down
a slope) (Payton et al. 2003, p. 357). Until the 1950s, very few African soil maps
existed that were based on actual surveys (FAO/Unesco, 1977, p. 1). The idea of a
map of the African continents soils based on regional soil surveys emerged at the
Commission de coopration technique en Afrique au Sud du Sahara (CCTA)s
Inter-African Pedological Service (Service pdologique inter-africain SPI)s first
Administrative Council meeting in Yangambi (then Belgian Congo) in 1953. At its
second meeting in 1955, the SPI Administrative Council recommended that a 1:
5,000,000 map for soil conservation and use be drafted in close collaboration with
the four regional committees southern, eastern, central and western Africa. This
decision stimulated preparation of new maps throughout the continent. Subsequent
meetings produced compromises among Belgian, French, British, Portuguese and
South African classification systems to reach agreement on a uniform map legend.
When produced in 1963, it was the first map ever drawn as the result of international
agreement, and made it possible to establish the relationship between pedogenesis
[soil formation] and the development of major soil units (FAO/Unesco 1977, p.
1). This map legend became the reference point for subsequent work. The idea of
a World Soil Map arose at the seventh Congress of the International Soil Science
Society in Madison, Wisconsin. The CCTA African soil map provided the base
from which the first draft of the Soil Map of Africa was produced in 1968 (FAO/
Unesco 1977, p. 2). In the same year the Organization of African Unity (OAU)
published its International Atlas of West Africa. Since the scales and much of the
data were the same, the OAU map was incorporated into the second draft of the
FAO/Unesco map. This 1971 draft became the final version of the Africa map of
the FAO/Unesco Soil Map of the World (FAO/Unesco 1977, p. 2).
The years of consultative preparation described above demonstrate the ex-
tent to which the FAO/Unesco soil map embodied the international consensus of
western scientific approaches to both soil identification and knowledge of African
soils. Finally published in 1977, this paper map is accepted as the appropriate
source of soil information for studies at a continental, regional or global nature
(Nachtergaele, undated). It has been the basis for all subsequent African continental
soil maps, including those involving simplifications and transformations, such as
the USDA maps using their Soil Taxonomy classification system, and the Russian
simplified 1: 15 million scale map of world landscapes, as well as electronic maps,
including the recent digital maps (For historical description of the evolution of
130
Kate B. Showers
electronic soil maps of the world see Nachtergaele, undated, pp. 23). Changes in
the maps have been largely due to data transformations, in which mathematical
relationships between two or more soil characteristics are used to estimate or predict
missing information (For discussions of pedotransfer functions and taxotransfer
functions see Nachtergaele, undated, p. 4). Nachtergaele admits that some of the
information contained in the [World Soils] map is of uneven quality, often outdated
and completely wrong (Nachtergaele, undated, p. 5). National and regional soil
mapping exercises have been funded by a variety of institutions using a range of
classification systems (USDA Soil Taxonomy, Orstrom, FAO and local systems),
as listed in Nachtergaeles appendix, Country Soil Maps of Africa.
However, Eswaran et al. (1992) argued that the FAO/Unesco maps lack of
data perpetuates misinformation about tropical soils. As shown in Figure 6.4, when
it was prepared, approximately 7 per cent of the African continent had been covered
by large or medium scale survey maps with some ground-truthing for accuracy; 38
per cent had coverage by reconnaissance maps that showed soils in relation to other
features such as climate, geomorphology or vegetation, and 55 per cent of the continent
was virtually unknown (FAO/Unesco 1977, p. 7). Information from a few specific,
detailed surveys was extrapolated to areas for which no information existed. These
best estimates, once displayed in map form on paper or with digital technology,
have the appearance of authority (See Figure 6.5 for example). Boardman (1998)
decries scientific myth making, in which specific and reliable data travel tortuous
routes and undergo transformations, making them incredible.
While Boardmans discussion centres on the extrapolation of carefully col-
lected measurements of soil erosion from a specific place to a regional, national or
continental scale, the same could be said for maps based upon similar processes.
For these reasons, people interested in specific places find local soil knowledge
more useful than internationally constructed soil maps reflecting international
consensus on the ways in which soils should be distinguished and classified (Nie-
meijer and Mazzucato 2003, pp. 404, 411). Similarly, those interested in making
general statements about soil over large areas recoil from the regional variation and
inconsistencies inherent in local soil knowledge.
A significant amount of effort has been put into comparing local soil classifica-
tion systems with those of western science. Since the systems being compared have
completely different origins and, often, purposes, the match is invariably imperfect.
It is extremely important to avoid the common assumption that western scientific
systems of knowledge are superior to other knowledge systems. Many studies by
western scientists have validated distinctions identified by, and conclusions drawn
from, local soil knowledge systems (WinklerPrins 1999, p. 153). Krasilnikov and
Tabor (2003, p. 203) point out that soil classification systems provide a common
means for talking about soils by simplifying a complex continuum into discrete
classes. For this reason, they argue, all soil classification systems are artificial, based
131
A History of African Soil
upon the information available and the beliefs and needs of the classifier. Each type
of information has its purposes; care must be taken to ensure appropriate applica-
tion. Although these systems should be understood and used as complimentary
perspectives from which to learn about the soil and its processes, the apparent
authority of western science with its generalised maps and specialised terminology
threaten the very existence of complex and detailed local soil knowledge systems
(Krasilnikov and Tabor 2003, p. 202).
132
Kate B. Showers
Figure 6.5. African soil types according to the FAO World Soil Resources map, 2003
133
A History of African Soil
3
African archaeology has been both under-funded and hindered by ideas generated by
archaeological work in Western Asia and Europe, as well as by theories about stages of
development. Vast areas of the continent have had little or no archaeological attention,
and important questions within archaeology have not been addressed. Like African soil,
the findings of African archaeology do not conform to experiences elsewhere, and have,
therefore, been either ignored or accepted with great difficulty. For general discussion of
these problems see Stahl 2005a, and for discussion related to specific topics, from hominims
to urbanisation, see various authors within Stahl 2005b.
4
Like all archaeological reconstructions, the detail of Olivers model of humanitys spread
is disputed. See Stahl 2005b.
134
Kate B. Showers
soil. Deshret was the name for red desert land. Three thousand years ago Egypts
soils had established commercial values: nemhura soils cost three times more than
sheta-teni soils (Krupenikov 1981 cited in Krasilnikov and Tabor 2003, p. 199). In
other parts of Africa (for which written records are sparse or absent and historical
linguistics methods have not yet been applied),5 ancient soil names have not been
documented and dates can be less clearly assigned to soil classification, value or
management systems. However, Central African oral tradition makes it clear that
the Bantu settlers arriving in the rainforest regarded knowledge as a key resource
(Guyer and Belinga 1995; Klieman 2003). They eagerly acquired the environmental
knowledge required for survival, as well as that for pottery production, from the
Batwa, the autochthons living in the rainforest (today referred to as Pygmies). Soil
knowledge was certainly a component of this information transfer.
Archaeologists have shown that first herding (cattle, sheep and goats) and,
much later, the cultivation of food plants, spread slowly across the continent, inter-
mingling with existing lifeways based on foraging, hunting and fishing.6 It is these
newer forms of sustenance that had the potential to influence soil properties, and
thus offer the opportunity to consider human beings as soil forming factors on the
African continent. Unlike patterns of land use evolution in western Asia/the Near
East and Europe (Stahl 2005a), the earliest food producers in Africa were mobile
pastoralists who left limited archaeological traces (Shahack-Gross et al. 2003), not
agriculturalists. As a stable and widespread way of life, pastoralism existed in the
Sahara and Sahelian grasslands by around 4500 bc (Gifford-Gonzalez, 2005, p.
188; Neumann 2005, p. 257). By 2000 bc they produced ceramics, which were
traded to hunter-gatherers (Shahack-Gross et al. 2003, p. 440).
While archaeological evidence clearly shows the development of African
pastoralism, documentation for the rise of agriculture is less clear. Neumann (2005)
usefully distinguishes between cultivation, any human activity that increases the
yield of harvested or exploited plants, and domestication, the genetic, morpho-
logical, and physiological changes of the plants resulting from cultivation and
conscious or unconscious human selection (Neumann 2005, p. 250). Cultivation
can be practised with wild plants as well as with domesticated plants, an important
concept when trying to understand African land use systems, and one that chal-
5
For discussion of the methods and applications of historical linguistics, see Klieman
2003.
6
for excellent review articles with comprehensive bibliographies see Stahl 2005b. It is
important to remember Kjekshus (1996) restatement of a main theme in Johnson and
Anderson (1988) that it is unhistorical to assert rigid classifications of African societies by
ethnicity or economic roles such as pastoralist, agriculturalist or hunter-gatherer, since these
societies were both socially dynamic and ecologically adaptive as they met survival challenges
(Kjekshus 1996, p.xxii).
135
A History of African Soil
sizes, but not necessarily increased soil disturbance (Neumann 2005, p. 266). Hoed
fields have rough surfaces that inhibit erosion, and hoed clods left in place serve as
a mulch, protecting the soil surface (Showers 2005, pp. 12, 14).
Causality between the spread of iron tools and agricultural diversification is
not clear. Where trees needed to be cut or other vegetation cleared or removed, the
acquisition of iron tools made agricultural production easier. However, in Central
Africa, it is argued, it was the introduction of a high-yielding, less labour-intensive
crop the banana that created the possibility for the development of metallurgy.
This more productive agricultural system could support non-agricultural specialisa-
tion (Klieman 2003, p. 99). Vansina (1990, p. 6) argues that it was the banana, not
iron tools, that enabled Bantu immigrants settled along rainforest river systems to
move deeper into the forest and colonise all of its habitats everywhere.7
Oliver (1991, p. 49) states that modern African cultures reflect the fact
that all of the life styles hunting, fishing, herding and agricultural production
coexisted, and persist. What is difficult to show is the effect these diverse and
complex land use systems have had on soil properties and soil bodies during the
millennia in which they have been practised. It can be argued that life styles centred
on fishing and hunting had little influence on soils. Although the East African
trade in ivory dates to at least 2000 years ago, and hunting was an integral part of
most African societies, hunting techniques other than the digging of pits in which
to trap animals did not disturb the soil (MacKenzie, J.M. 1988, pp. 63, 77, 148).
In contrast, livestock management and crop production have potentially signifi-
cant consequences for African soils. The basic concepts of dispersed grazing with
transhumance (seasonal movement of stock to a different region) and crop-fallow
rotational cropping systems have been applied with variation over much of the
continent for millennia. Even where there have been traditions of urbanisation
in Africa, studies have shown that these settlements were always linked to larger
landscapes; sub-Saharan iron-using urbanised people engaged in mixed farming
date to around the first millennium ad (LaViolette and Fleisher 2005, pp. 332,
327; Oliver 1991, p. 93; Lewis and Berry 1988, p. 29).
Archaeologists have begun to study the prehistory of land use systems.8 Part
of the challenge of this work is to overcome such soil problems as acidity that cause
7
While discussion of the Bantu migration is beyond the scope of this chapter, it should be
kept in mind that, as Childs and Herbert (2005, p.281) point out, Bantu-speaking peo-
ples were borrowers rather than innovators; they acquired knowledge of ironworking, seed
agriculture, and animal husbandry from their non-Bantu-speaking neighbors. For critical
discussion of the Bantu migration, see Vansina 1990, Klieman 2003 and Eggert 2005.
8 Neumann (2005, pp.2501) provides a useful history of the archaeology of African plant
food production, noting the subjects very recent origins.
137
A History of African Soil
a. Livestock Management
Africa has experienced ancient traditions of dispersed or extensive grazing, since
pastoralists move according to the needs of their stock for pasture and water. Cattle
were tended on the continent before sheep and goats, and in Egypt before the Sahara,
but by about the fifth millennium bc a successful cattle-based pastoral economy
existed in the central Sahara. This ended during the arid years from approximately
35502550 bc as pastoralism shifted to the south and east. Cattle, sheep and goat
pastoralism (the Saharan triad) first appeared in East Africas Lake Turkana basin
around 20502550 bc. Less archaeological evidence exists for southern Africa,
but it seems that only sheep arrived in south-west South Africa in about the first
millennium ad (Gifford-Gonzalez 2005, pp. 203, 204, 206). The existence of
pastoralism in the absence of agriculture has been explained in terms of a system
that could respond successfully to climate change (Gifford-Gonzalez 2005), and
as a response to Africas rich environmental resources, particularly in the savanna,
138
Kate B. Showers
wet (Tanner and Mamaril 1959; Gradwell 1968; Twerdoff et al. 1999). Hooves dig
deeper into wet soils, compressing and remoulding or puddling them (Gradwell
1968). Damage under these conditions could reduce pasture growth, and could
be irreversible. The severity of compaction depends upon the intensity of grazing,
soil type, and vegetation (Warren et al.1986; Krenzer et al. 1989; Twerdoff et al.
1999). Alternating periods of grazing with nongrazing allows some, if not complete,
recovery (Gradwell 1968; McCarty and Maurak 1976; Haynes and Francis 1993).
Studies have shown that compaction occurs only in surface soils (Krenzer et al.
1989), and dense vegetation can mitigate this consequence of grazing. Although
soils in grazed pastures may be permanently denser or more compacted than those of
nongrazed areas, this does not necessarily affect plant growth (Tanner and Mamaril
1959; Gradwell 1968). However, water enters a compacted soil with greater dif-
ficulty, and compacted soils can hold less water. Compacted soils can, therefore,
be drier than non-compacted soils, so that plants growing on them are more likely
to experience water stress (Gradwell 1968).
Livestock play a role in nutrient cycling as they eat and then deposit manure
and urine. This is discussed in detail below, under savanna systems of agricultural
production. Powell et al. (1996, p. 145), working in the semi-arid Sahel of West
Africa, argued that cattle grazing on rangeland represents a closed cycling system,
but as livestock producers increasingly settle, corralling of range-grazed livestock
on crop land could create imbalances.
Because in many African pastoral systems livestock were herded to different
locations in the landscape for food and water at different times of the year, what
were recognised in the late twentieth century as conservation range management
principles were being followed (Pendleton 1982; Coupland 1979). There is a pos-
sibility of some intensely grazed areas having permanently compacted soils, which
affected soil-water relations. However, it is reasonable to assume that most African
pastureland was unstressed by traditional pastoralists. This is not to say that there
are no examples of soils disturbed by livestock management. River crossings and
paths certainly had compacted soils, and where large herds of livestock were concen-
trated (such as the settlements of the Zimbabwean plateau in the eleventhfifteenth
centuries ad, particularly Great Zimbabwe), all the ills associated with over grazing
and trampling undoubtedly occurred.
b. Agricultural Production
Nye and Greenlands (1960) classic study The Soil under Shifting Cultivation assesses
the effects of crop-fallow rotation systems in both the evergreen and semi-decidu-
ous high forests and the tall tussocky grass and tall and short bunch grass savannas
of West Africa. Samples were taken and literature reviewed of soil and vegetation
systems in the mature fallow state, immediately after clearing and burning, and in
the succeeding three years of crop production. At the time this study was made,
140
Kate B. Showers
9
Vansina (1990) and Klieman (2003) note that recent research in the Central African rainforest
points to greater diversity than previously imagined. See discussion of African soil maps.
141
A History of African Soil
On the acid, highly weathered soils of the West African moist evergreen
forest, cropping could only be supported for one and a half to two years; a ten-year
fallow was required to restore fertility. The relatively more fertile soils of the drier
semi-deciduous forest could support crop production for three years, followed by
an eight-year fallow. It was only early in the first year that the soil surface was af-
fected by planting crops in either type of forest. During that time, in both systems,
the soil of the cleared space was bare. With the first rains, seeds were planted in the
surface soil (which was rich in organic matter and worm castings), with a digging
stick causing minimal soil disturbance. Annual grains and perennial crops were
planted as an intercrop. Until the crops grew, the soils surface was exposed to heavy
rains that could destroy the structure of both exposed aggregated soil particles and
worm castings, and ash could float away in runoff. Sheet or rill erosion could also
occur. However, within four weeks the crop plants (and, likely, weeds) had produced
a canopy that covered and protected the soil surface. In subsequent years the soil
was never bare, as perennial crops were harvested and there was no replanting (Nye
and Greenland 1960, pp. 3, 4).
Since animals were not kept by forest dwellers, the non-soil sources of soil
fertility were, primarily, the decomposition of organic matter and additions from
rainfall, dust, lightning, and the fixation of atmospheric nitrogen by microorgan-
isms. The major losses from the system were considered to be due to leaching
of nutrients below the rooting zone, removals in run-off water, the volatilisation
of nitrogen by microorganisms, and of nitrogen and sulphur through oxidation
when organic matter was burned. Nye and Greenland concluded that forest crop-
fallow systems were nearly a closed cycle and, therefore, efficient and sustainable
(Nye and Greenland 1960, pp. 43, 44). The system as described does not suggest
modification of soil properties beyond temporary changes in the soils chemistry
(plant nutrients) and microbial populations. More recent international agroforestry
research reviewed by Kass et al. (1999) supports these conclusions.
The Central African rain forests (modern Rwanda, Burundi, Uganda,
Democratic Republic of Congo, Cameroon, Equatorial Guinea, Gabon and
Congo Republic) have been occupied by human beings for millennia (Sayer et
al. 1992, p. 43). The earliest evidence is for isolated groups of stone tool-using
hunter-gatherers living on hilltops and near running water from 38,05033,050 bc,
during the Ndjilian humid phase, when Gabons forest refuges were first formed.
This was followed by 23,000 years of lower rainfall and temperatures associated
with the Leopoldian era climate shift, when forest cover shrank and savannas ex-
panded. During the subsequent humid Kibangian era (from 10,050 bc), the forest
expanded to its current size, and residents produced small, less heavy stone tools
and engaged in medium-distance trade (Klieman 2003, pp. 37, 40). The meagre
evidence for subsistence suggests a reliance on wild food. However, at archaeologi-
cal sites in the Ituri rain forest dating back 10,000 years, there are remains of the
142
Kate B. Showers
oil-producing plant Canarium schweinfurthii, and later sites containing oil palm
(Elaeis guineensis). The presence of these plants has been cited as evidence of pos-
sible early plant management (Casey 2005, p. 234). However, Casey points out,
modern foragers use these plants, and lightning strikes and windfalls can open up
spaces in forests providing the light required for oil palm survival (Casey 2005,
p. 234, 235). Heeding Neumanns caution about assuming agricultural activities
from evidence of cultivated plants, it is not reasonable to conclude that these were
farming communities.
There is evidence of soil knowledge among ancient rainforest residents.
Giles-Vernicks (2002) account of the knowledge system of the Mpiemu hunter-
gather-farmers of the Central African Republics Sangha River Basin, makes clear
that local environmental knowledge, including soil knowledge, was part of their
identity. Central to becoming a person in Mpiemu culture was the transmission
of knowledge from one generation to the next. Giles-Vernick could only trace the
Mpiemus presence back 150 years, but her informants claimed an ancient origin
for their knowledge, a time when lives lived in grasslands combined with those of
the forest (Giles-Vernick 2002, pp. 49, 56, 57).
Vansina (1990) describes the process of multi-generational experimentation
in West African forests that began 5000 years ago, leading to more sedentary lifeways
and, ultimately, the agricultural systems of Bantu speakers based on yams, gourds,
castor bean, black-eyed peas, and the Voandzeia groundnut (Klieman 2003, p. 41).
These agriculturists began to move into the Central African rainforest along rivers,
and by the middle of the second millennium bc had established many communi-
ties in the western equatorial forest. These settlements were involved with trade
networks that carried ceramics by the fourth century bc , but it isnt until 500 bc
that there is evidence of slash-and burn agriculture (Klieman 2003, pp. 40, 55).
Sayer et al. (1992), concurring with Vansina (1990), make clear that all parts of
Africas tropical rainforest are now used by local residents, and that there is no area
of forest that has not had a long history of human use.
According to Vansina (1990) the selective use of plants and animals by
dispersed hunter-gatherers did not significantly alter rainforest dynamics, but the
arrival of settlers who farmed did. Farming populations were kept low, he asserts,
because large populations cut down so many trees that the supportive habitat was
destroyed. Vansinas concern was for flora and fauna, and not soil. However, aside
from the (unstudied) consequences of ceramics production, there is no evidence of
soil disturbance but none has been sought. Given Nye and Greenlands evidence
of crop-fallow rotation systems in moist evergreen forests producing only temporary
changes in soil properties, and noting the low density of human populations in
the past, it could be argued that with respect to crop production, moist evergreen
forest dwellers in the past were nothing more than participants in, and users of, an
143
A History of African Soil
almost closed nutrient cycling system, and did not significantly alter soil properties
or soil bodies.
This might not have been the case in the drier woodlands of central Africa.
The Bemba, who inhabit northern Zambias Miombo woodland, named their crop-
fallow rotation system chitemene. Chitemene involves burning on a new field not
only the trees cut in the process of clearing, but also branches and brush cut from
adjacent land. The resulting intensive burning heated the top 2.55cm of soil, with
the result that soil structure improved (Nye and Greenland 1960, p. 71).
Soil properties were changed in the chitemene system for the short and
medium term. Along with the expected initial transfer of nutrients from vegetation
to soil, Stromgaard (1986) found that burning increased the sandy soils ability to
retain cations (cation exchange capacity, or CEC) for as long as 16 years after the
burn, when the chitemene cycle could begin again. While the amount of basic cations
(calcium, magnesium and potassium) retained in the soil might be considered to
be low, they were essential for regeneration of the fallow vegetation. Stromgaard
(1986) also documented the encroachment of grasses during a crop/fallow cycle,
noting that the dense root systems prevented cultivation. Invasive grasses of the
Hyparrhenia family secrete a toxin that suppresses the growth of nitrifying bacte-
ria, thus reducing the amount of nitrogen available in the soil and retarding the
regeneration of woody species (Stromgaard 1986, p. 104).
Stromgaard (1986) argued that the area cleared for a field and the sur-
rounding land from which tree branches were cut must be viewed as components
of a system consisting of an intensively farmed infield and an outfield exploited
at low intensity (Stromgaard 1986, p. 97). The chitemene system concentrates
nutrients from a given areas vegetation in a smaller space, and improves that soils
ability to retain them. It also facilitates the growth of grass plants which affect the
microbial environment. If these changes persisted over time, then chitemene could
be considered to transform the properties of the soils on which it was practised.
This form of cultivation is practised in most parts of the plateau in north-western
and north-eastern Zambia (Moore and Vaughn 1994).
Savanna Systems. Although savanna vegetation is far more extensive than moist tropi-
cal forest, human relations with savanna soils are less clear. By definition, savanna
soils occur where rainfall is too low to support a closed-canopy forest (See Figure
6.2). They are, therefore, less weathered and tend to have higher pHs and nutrient
supplies. Perhaps surprisingly, soil organic matter levels of savanna topsoils were
found to be lower than those under forest cover.10 Rotational use of savanna soils
for cropping has followed some of the principles and practices of forest systems,
10
Plant roots and associated microbial communities, not aboveground growth, are the major
source of soil organic matter (Bolinder et al. 1999).
144
Kate B. Showers
but there are significant divergences. As in forest systems, low intensity burning
did not affect soil temperatures below the surface 5 cm. of soil, but intense burns
did (Nye and Greenland 1960, p. 11). The annual low intensity burning of the
savanna was a mechanism for returning nutrients to the soil but, since grasses do not
increase their store of nutrients year by year as forests do, when savanna vegetation
was burned, fewer nutrients were deposited on the soil surface than when a field
was cleared from forest (Nye and Greenland 1960, p. 36). The rapidity of annual
burns and its positive effects on vegetation have been noted by recent researchers
(Laris 2004; Wardell et al. 2004).
Parallels cannot be drawn between the values of mature forest and mature
savanna. When cleared, old savanna not only had distinctly lower nitrogen levels
than mature forests, but also less than savanna of two to six years. Nye and Green-
land argued that because the long dry season meant that only annual crops could be
planted in savanna regions, the soil was left bare each year after harvest. Additionally,
surface soils had to be disturbed for planting because of the need to remove grass
roots, and remained uncovered until crop canopies developed. Finally, plant cover
was slow to establish when the fallow period began because of a scarcity of suckers
and established seedlings (Nye and Greenland 1960, p. 124). This lack of cover,
they argued, exposed savanna soils to erosion by water and wind. If eroded enough
to expose subsoils containing plinthite, a very hard surface could form.
More recent research suggests that at the very least, dry-season soil was covered
by weeds and unharvested plant parts (stover). In the moist high-grass savanna, as
in the evergreen moist forest, livestock was limited by tsetse flys trypanosomiasis,
so manure was not part of the nutrient cycling system; soil fertility was maintained
exclusively by fallow management. However, in the drier tall and short bunch grass
savanna, where there were no tsetse flies, herds of cattle, sheep and goats were sig-
nificant components of soil fertility management (Nye and Greenland 1960, p. 4).
Hoffman et al. (2001, p. 268) described precise and careful management of post
harvest stover and weeds for livestock grazing during the dry season in semi-arid
northwest Nigeria. Grazing stock converted high carbon organic matter (stover) into
more plant-available forms of nutrients and deposited them as urine and manure.
The stubble mulch resulting from grazing protected the soils surface. Fields in the
study area (Kano close-settled zone) were only disturbed and bare at the onset of
the rainy season, when roots had been dug and burned. Both resident livestock
and those of pastoralists grazed the fields. So valued is manure for soil fertility
maintenance that farmers in semi-arid West Africa have traditionally arranged with
migratory pastoralists to corral their herds for a fixed number of nights on their
fields (Williams 1999, pp. 17, 18). These arrangements were especially important
for farmers without large numbers of livestock.
Detailed research presented in a series of papers by Powell et al. (1996,
1998) and Ikpe et al. (1999) investigated the role of livestock in maintaining soil
145
A History of African Soil
fertility in the Sahel. Their experiments showed that while manure was important,
it was the deposition of urine that had the most significant and long-term effects
on subsequent crop yields. The research was carried out on a Labucheri soil, which
is very low in organic matter content, has predominantly kaolinitic clays (very low
ability to supply nutrients), and is poorly buffered (cannot easily resist acid-base
changes). The addition of urine caused two major nutrient changes in this soil.
As expected, levels of nitrogen in various forms increased. Surprisingly, levels of
phosphorus also increased. Since there is virtually no phosphorus in urine, the
researchers theorised that when the urine raised soil pH (the soils became less acid)
during the first week, phosphorus dissolved out of the aluminium-iron complexes
in the kaolinitic clays and became available to plants (Ikpe et al. 1999). Technically
speaking, phosphorus was released when the structure of soil particles was ruptured.
This process can be understood either as nutrient cycling shifting phosphorus
from unavailable to available forms in the soil or destructive mining, since the
release of phosphorus involves the destruction of individual soil particles. Powell et
al. (1996) argue that cattle grazing a rangeland represent a closed nutrient-cycling
system, but when grazed on rangeland by day and pastured on cropland by night,
livestock are the principal vectors of nutrient transfer across the landscape (Powell
et al. 1996, p. 148; Powell et al. 1998, p. 260).
The Gourmantch people in the southern Sudano-Sahelian (900 mm mean
annual rainfall) and more northern Sahelian (600 mm mean annual rainfall) regions
of eastern Burkina Faso were found to have complex understandings of nutrient
cycling systems, the role of organic matter, and the relative importance of water and
soil fertility to crop production in a region with highly variable rainfall amounts
and distribution (Niemeijer and Mazzucato 2003). The area consists primarily of
flat to gently undulating peneplain with occasional lateritic hills. Most of the soils
have low fertility and low levels of organic matter; many are shallow, gravelly or
have lateritic formations. They can also crust, seal or become hard. The vegetation
ranges from savanna woodlands in the south to tree and bush savannas in the north.
The Gourmantch believe that the landscape is shaped by erosion, sedimentation
and the transformation of organic matter. They think that erosion could be a source
of soil fertility, as well as create gullies that could, over time, become a valley. They
also think that few cultivable soils are inherently good; water relations not soil
chemical or physical properties determine a soils utility. A soil in a water collect-
ing spot would be good in a drier year, and bad in a wetter one. No soil would be
good in a drought. Soil fertility is understood to be a function of organic matter,
and distinctions are made based upon its source, location (on top of, or in, the
soil) and degree of decomposition. Darker soil colours are associated with organic
matter so decomposed that it is no longer visible. Gourmantch farmers believe that
plants consume organic matter from the soil, and that repeated cropping reduces
soil fertility, making soils lighter in colour. They also realise that too much organic
146
Kate B. Showers
matter or manure limits crop production. While livestock are appreciated for the
manure they produce, their trampling is associated with decreased ground cover
and subsequent erosion, as well as with soil crusting and compaction. Because the
Gourmantch farmers see a relationship between organic matter additions and
soil fertility, they believe that there cannot be an irreversible loss of soil fertility.
In dry years, when vegetation is sparse, soil organic matter levels decline, but in
wetter years with more luxuriant growth, the levels increase again. Farmers, they
believe, can work within this continuum of vegetation supply (and manure) to
maintain soil fertility.
d. Architecture/Urbanisation
Whether for building one-roomed house components, large mosques, markets or
walls, soil as brick or plaster has been used for centuries in many parts of Africa
(for photographs and descriptions of regional architectural styles see Denyer 1978).
Not all soils are suitable for construction. It is only those with high concentrations
of certain types of clays that make good bricks and plaster. This suggests that for
as long as people have used soil for building, there has been a basic knowledge of
soil textural properties, and an ability to distinguish one soil type from another.
The excavation of these soils in sufficient quantity for construction undoubtedly
resulted in local soil disturbance. The construction of one households buildings,
however, was not as consequential as the creation of urban areas that not only
disturbed soils they covered, but also stimulated extraction from and use of
larger landscapes.
Although it has been described as the least urbanised continent ,11 Africa
actually has a long tradition of cities and towns. Since the architecture in many
African settlements consisted of structures made from easily degraded materials
such as straw, earth, or small trees, they were not easily preserved or recognised.
With or without the monumental architecture associated with classical definitions
of urbanity, concentrated settlements of human populations did exist in Africa,
and did affect the larger landscape, including soils.
The Nile valley has a history of urbanisation dating from 3500 bc, and urban
areas have existed along the Mediterranean Coast of Africa from 1200 bc (Lewis
and Berry 1988, p. 97). Large-scale settlements existed in West Africas Inland
Niger Delta from the first millennium ad. One such place, Jenn-jeno, began as a
seasonal settlement in about 250 bc, covered 12 hectares by 100 ad, and grew to
a permanent village of daubed pole-and-mat houses measuring 25 hectares by ad
400. From ad 400 900 Jenn-jeno occupied 33 hectares, with a cautiously esti-
mated population of 3,2006,400 in town, and a further 6,70013,400 in nearby
settlements. A massive wall of sun-dried cylindrical bricks enclosed the 33-hectare
site (LaViolette and Fleisher 2005, pp. 328, 334). Continuous urban settlement in
the Chad Basin dates back to the ninth century ad. Ngazagarmo, the capital city
from the fifteenth to early nineteenth century, was at its height in the seventeenth
century. It covered about 1,500 ha and contained a quarter of a million people
11
Until the 1980s, Northern Hemisphere conventions about and definitions of cities
and urban areas largely excluded Africas urban reality and past from consideration. For
discussion, see LaViolette and Fleisher 2005.
149
A History of African Soil
(Denyer 1978, p. 35). In 1658 Ngazagarmo reportedly had four mosques and 666
roads cleared and widened (Denyer 1978, p. 35).
In the Sahara desert a chain of cities stretched from Tekrur in the west
through Audoghast, Timbuktu and Gao to Agadez in the east from the tenth to
the fifteenth centuries (Denyer 1978, p. 31). Further south, along the northern
limits of the tsetse fly, a second line of cities developed that included Segou, Djenne,
Ouagadougou, Katsina and Kano. From this line of Hausa urban areas south,
many towns had extensive walls enclosing not only buildings but also open land
(Denyer 1978, p. 33). The Yoruba cities of Western Nigeria, which probably date
from the fifteenth century, had town centres consisting of a palace, a market and
a temple with its sacred grove of trees (Denyer 1978, pp. 35, 36). Two main roads
divided the city into quarters. Houses were built close together in family units,
surrounding a well, rather than along a defined street (Lewis and Berry 1988, p.
350). According to Lloyd (1967, p. 220), Ibadan and Abeokuta were surrounded
by hamlets from which the farming community commuted, while Ijebu Ode was
surrounded by more than 100 permanent villages. Lloyd noted the antiquity of
the villages in the mid-1960s.
On the East African coast, the town of Rhapta (in modern Tanzania) was
reported by Ptolemy to have grown from an emporium to a metropolis by ad
300 (Chami 1996, p. 16). A series of towns and villages based on local, regional
and long-distance trade dating from the eighth to the eighteenth centuries dotted
the coast from modern Somalia to Mozambique (LaViolette and Fleisher 2005, p.
339). Building materials ranged from stone to earth-and-thatch, and populations
from more than 5,000 to 15,000 people (LaViolette and Fleisher 2005, p. 340).
The towns tended to be arranged around a well or market, not along streets, and
household units included garden areas for domestic food production (Lewis and
Berry 1988, p. 351). The merchants in these towns controlled local, regional and
international trade, and thus the towns were connected to the larger landscape and
inland settlements by paths and roads.
In southern Africa, cattle-keeping and farming populations on the Zimba-
bwean plateau built settlements from the first millennium ad. Buildings were made
from stone and from coursed earth, called daga. In time, political states developed,
based upon cattle-keeping and the gold trade, with major settlements at places
such as Mapungubwe, Khami, Danangombe, and Great Zimbabwe (Beach, D.N.
1984). These thirteenthfifteenth-century cities are renowned for their monumental
stone structures, but daga housing also existed (LaViolette and Fleisher 2005, pp.
2367). North of the Orange River in modern South Africa, eighteenth-century
Tswana people lived in concentrated settlements while operating an extensive land
use system. European travellers in the early nineteenth century reported that the
Tlhaping (a Tswana speaking group) capital was the same size as the colonial set-
tlement of Cape Town (Jacobs 2003, p. 45).
150
Kate B. Showers
No matter what their construction materials, African urban areas affected the
soils they occupied as well as the surrounding landscape. Soil surfaces were covered,
roads and paths were created, water was drawn, used and disposed of, and waste was
discarded. The result was soil compaction, disruption of soil water relations, and
pollution. Increased demands were placed on food-producing areas to supplement
urban gardens. If soil was used for construction, quarries had to be dug, and when
a building was no longer used, the soil had to be either reused or disposed.
e. Mining
More obviously disruptive of soil than agriculture, livestock production or urban
areas is mining. Holes or tunnels in the ground or in stream banks are dug to extract
desired minerals or soil types. Not only does the removal of the desired material cause
disruption of the soil, but purification can result in large quantities of discarded
waste. For millennia Africans have mined iron and copper for tools, ornaments,
currencies and ritual objects, and salt and clays for nutritional, medicinal or ritual
use. Small- and large-scale mining and metal working centres existed throughout
Africa, producing ore and finished products for local use as well as for regional
and long-distance, international and intercontinental trade networks. Mining was
a major source of wealth for African empires and kingdoms (Trore 1994 cited in
Hilson 2002, p. 154).
Metallurgy. Minerals can be collected from the soils surface, retrieved from stream
beds, or excavated from beneath the ground. Once mined, mineral ores must be
separated (smelted) from non-ore materials by heating. Smelting results in some
pure metal and much larger quantities of waste, called slag, that is usually deposited
on the soil surface. To produce finished objects, the metal must be forged heated
and worked. Since each of these processes has potential to affect soils directly or
indirectly, it is important to have an idea of their nature and extent.
Iron, copper and gold were the primary metals mined and processed in
Africa lead and tin to a lesser extent (Childs and Herbert 2005, pp. 281, 282).
The centres of mining and skilled craftsmanship developed in areas rich in ores,
from which regional and global trade networks spread. Among Africans iron was
the most commonly used metal (both for domestic and ritual use), and in many
societies copper was more valued than gold. Copper-poor West Africa imported
it across the Sahara in about the twelfth century ad (Childs and Herbert 2005,
p. 290). At approximately the same time, gold, whose primary demand was non-
African, was being exported to the Arab and Indian Ocean worlds. The substantial
trade with Europe came later (Childs and Herbert 2005, p. 287). Hilson (2002, p.
153) suggests that gold was first mined intensively from Archaen greenstone belts
in modern Zimbabwe, Botswana, Tanzania, Mozambique and South Africa.
151
A History of African Soil
African metal workers knew a great deal about the properties of metal. They
produced not only wrought iron and cast iron, but also low- to medium-carbon
steel. Steel production requires the manipulation of conditions inside smelting
furnaces (for descriptions of the great array of furnaces and unique African furnace
design that eliminated the need for a bellows see Childs and Herbert 2005, p. 284).
In southeast Nigeria, the lost-wax casting process was highly perfected.12 Central
and southern African smiths made copper and iron wire by both hammering and
drawing (Childs and Herbert 2005, p. 286).
The development of Africas sophisticated metallurgy traditions, like those of
pastoralism and agriculture, does not conform to patterns identified by European
archaeologists. There was no Bronze Age in Africa, and evidence points to the smelting
of iron before copper (Childs and Herbert 2005, pp. 227, 228). Radiocarbon dates
of wood (the primary fuel in smelting) from scattered sites around the continent
going back to the second millennium bc, confirm metalworking to be an ancient
African activity. Rather than finding evidence of iron working as an imported skill,
calibrated radiocarbon dates from about 800400 bc at archaeological sites in Ni-
ger, Nigeria, Gabon, Cameroon, the Central Africa Republic, and the Great Lakes
supports the possibility of independent invention. However, there are not enough
data to chart the spread of ironworking across the Africa (Childs and Herbert 2005,
p. 280). Everything to do with metallurgy had soil consequences.
Mines. Most obviously, metals must be removed from the soils containing them.
Knowledge of soil properties both the staining of surface soils and the kinds of
vegetation supported was used by ancient prospectors to locate ore deposits (Childs
and Herbert 2005, p. 282). Since iron is so common in Africa from deposits
on individual soil particles (as mentioned earlier) to concentrations of ore there
should be no surprise that it was the most widely and continuously mined mineral.
Ironworkers were able to distinguish between lower and higher grade ores, and only
used oxides such as haematite, magnetite and limonite (Childs and Herbert 2005,
p. 282). The greatest number of less frequently occurring copper deposits are in
central and southern Africa. With the exception of Zambias Copperbelt that was
identified with twentieth century prospecting technology, almost all of the known
sub-Saharan copper deposits were worked by miners in the pre-colonial era (Childs
and Herbert 2005, p. 282). Although gold is widespread throughout the continent
in both alluvial and reef forms, the earliest archaeological evidence for its use is not
until the seventh and eighth century ad at Jenn-jeno (West Africa) and in elite
burial sites at Mapungubwe, near the Limpopo River in southern Africa, dated to
12
Lost-wax or cire perdue casting involves making a model of the object to be cast in wax,
then enveloping it in clay. When the wax is melted in an oven, the metal is run in to take
the exact shape of the lost wax (Childs and Herbert 2005, p.286)
152
Kate B. Showers
approximately the twelfth century ad (Childs and Herbert 2005, p. 282). A lot of
southern African gold came from alluvial (river) deposits and was panned (Tlou
and Campbell 1984, p. 42).
Establishing the history of mining has been complicated by the fact that
modern mines have been dug on top of ancient mining sites. The most common
type of mine was a shallow pit or shaft dug with hoes or digging sticks. In some
locations, mines consisted of holes dug into hills, creating caves, which were used
until they collapsed (Tlou and Campbell 1984, pp. 42, 43). In central Africa, some
open-cast copper mines were very large. A pit in Katanga, Democratic Republic of
Congo was of a mile long and 6001000 feet wide, which compares favourably
to modern pits made with steam shovels. In nearby Kanshani, Zambia, an area
7,000 yards in diameter had been excavated. Surveyors have calculated that over
the centuries of mining, thousands of tons of malachite were removed (Childs and
Herbert 2005, p. 283). A lot of gold mined came from alluvial (river) deposits and
was panned (Tlou and Campbell 1984, p. 42).
Deep mines were required for both copper and gold in central and southern
Africas hard rock areas, particularly in Zimbabwe and South Africa. By ad 750,
specialised mining communities had developed in southern Africa and underground
mines were dug. Gold mining is thought to have begun in Zimbabwe around ad
900, and in northeast Botswana soon after (Tlou and Campbell 1984, p. 42).
In preparation for digging underground mines, fires were set to fracture
surface materials, followed by digging shafts with iron picks and hammer stones
(Childs and Herbert 2005, p. 283).
The deepest shaft found in Botswana measured 26 metres, a seventeenth-
century-ad Zimbabwean copper mine had shafts that extended down 60 feet, and
a gold mines descended 180 feet (Tlou and Campbell 1984, p. 42, Childs and
Herbert 2005, p. 283). Shaft depths were limited by an areas water table there
was no ancient technology for pumping. Once shafts had been dug, miners de-
scended by rope to dig ore, which was sent back to the surface in baskets (Childs
and Herbert 2005, p. 283).
Smelting and smithing. Along with mining, large-scale iron production centres
existed throughout the continent. In West Africa, the best known are the Middle
Senegal Valley, Futa Jalon in Guinea, Mema in Mali, Yatenga in Burkina Faso,
the Bassar region of western Togo, Hausa-speaking regions in Niger and northern
Nigeria, and the Ndop Plain of Cameroon. There were also major centres in Cen-
tral Africa (the Batk plateaux of Gabon and Congo), the Great Lakes region of
Rwanda/Burundi, Tanzania, and Malawi; the Shona-speaking region of Zimbabwe;
and Phalaborwa in South Africas eastern Transvaal (Childs and Herbert 2005, pp.
2834). The technology of iron smelting reached Botswana approximately 2000
years ago (Tlou and Campbell 1984, p. 42), and has existed in Kondoa Irangi,
central Tanzania, for a thousand years or more (Mapunda 2003). Evidence has been
153
A History of African Soil
f. Medicine
Geophagy, the eating of earth, has been common in Africa for centuries for me-
dicinal and ritual purposes. Clays from Uzalla, Nigeria that are sold widely in
154
Kate B. Showers
markets across West Africa, have a mineral composition similar to clays used in
the commercial pharmaceutical Kaopectate, the commonly used anti-diarrhoeal
remedy (Aufreiter et al. 1997, p. 293). Geophagy has been widely documented
in pregnant women, and theorised as being a source of iron (Abrahams 1997). As
well as providing a nutritional supplement, clays can also help with detoxification.
Eight samples of edible clays from Gabon, Kenya, Nigeria, Togo, Zambia and the
Democratic Republic of Congo (former Zaire) were analysed for their contents of
calcium, copper, iron, magnesium, manganese and zinc and detoxification capacities.
The samples were found to be able to provide nutritionally significant amounts,
and could, therefore, serve as a mineral supplement (Johns and Duquette 1991,
pp. 450,454). In addition, because they were all kaolin clays, they could serve the
same function as Kaopectate. A clay recovered from an archaeological site occupied
by ancestors of Homo sapiens was found to have nutritional and detoxification
capacities that were identical to those of edible clays used in Africa in the 1990s
(Johns and Duquette 1991, p. 455).
Very specific medicinal mining of clays has, therefore, been carried out since
the beginnings of human history. The extent and significance of this not been as-
sessed, perhaps due to the taboo which industrial societies attach to the worldwide
historical (and modern) practice of eating earth.
because it creates channels in fields in which water can collect or flow, reduces soil
organic matter levels, weakens aggregated soil particles, compacts subsoil to form a
dense layer (plough pan), can physically shift soil particles downslope, and results
in a more exposed soil surface than does hoeing.13 European livestock management
resulted in overstocked pastures and the concentration of animals around corrals
and watering holes, which caused soil compaction (see discussion of compaction
above under African grazing systems). These land use practices were associated with
soil erosion (Hailey 1938, p. 1080; Tempany 1949, p. 1; Jacks and Whyte 1938,
p. 71; Ross 1963, p. 12; McIlwane 1941 in Nyampafene 1987, p. 2). As Africans
adopted European practices, erosion began to occur on their land as well.
Soil erosion in Southern Africa cannot be discussed in isolation from the
great disturbance in the landscape caused by the arrival of new groups of land users
from Europe who perceived much of the region as unused or empty, and claimed it
for their own purposes. As the number of settler farmers increased, the land avail-
able to Africans decreased. Treaties signed in the late nineteenth century effectively
transferred most of the land used by the Swazi people in the Transvaal Republic
(north of the Vaal River in modern South Africa) to settlers (Swaziland Col.An.
Rept. 1956, p. 17), and the rich farmland of the Basotho (modern Lesotho) to the
Afrikaner farmers of the Orange Free State (modern Free State Province, South
Africa) (Germond 1967, p. 308). Large blocks of land with the highest rainfall in
the Bechuanaland Protectorate (modern Botswana) were declared Crown Lands
available for European settlement (Tlou and Campbell 1984, p. 181). Laws passed
in South Africa in 1913 and 1936 crowded the African population onto 13 per cent
of the territory, giving the tiny European population control of the remaining 87 per
cent of the land (Platzky and Walker 1985, p. 92), including the most favourable
agricultural regions of the country. In Southern Rhodesia (modern Zimbabwe),
Native Reserves were formally created for Africans in 1913 on marginal land with
infertile soils and low rainfall (Mpofu 1987, p. 2), again leaving the more favoured
land in the hands of European settlers.
The result was the thorough disruption of indigenous land use systems.
Although details and dates vary from nation to nation, the overall form is of systems
that had supported societies for generations and centuries, if not millenia, had their
land bases seriously reduced. Indigenous land use systems were extensive, requiring
large amounts of land. When confined to a smaller area, the African systems no
longer functioned well (Basutoland Dept. Ag. An. Rept.1930, p. 9; Nyampafene
1987, pp. 1, 2; Swaziland Col. An. Rept. 1946, p. 14; Swaziland Col. An. Rpt.
1950, p. 7; Swaziland Col. An. Rept. 1952, p. 52; Weinmann 1975, p. 200). First
13
For references and discussion of the consequences of ploughing on rangeland and of ero-
sion associated with ploughing see Showers 2005, pp.1922.
157
A History of African Soil
there was no surplus, and then not enough for self-sufficiency. This process resulted
in increasing and inevitably unbearable pressures on the landscape; pressures
that resulted in the cultivation of marginal land, overuse of pastures, deforestation
and soil erosion. In many areas accelerated soil erosion had been an unknown
and unnamed concept. By the end of the nineteenth century, overcrowding or
over-population of people and their livestock was identified as the major cause of
erosion in most countries of southern Africa. A land problem created by the arrival
of European settlers led to the erosion problem.
The earliest discussion of accelerated soil erosion occurred in South Africas
Cape Colony (Beinart 1984, pp. 54, 55). Apprehension about soil conditions was
not unique to this colony. In the adjacent Protectorate of Basutoland (modern
Lesotho), missionaries and government officials had expressed particular concern
about gullies along roads and paths, and on government and missionary land
(Showers 1989; Showers 2005). Further north in the colony of Southern Rhodesia
(modern Zimbabwe) and the Nyasaland Protectorate (modern Malawi), officials
worried about erosion on European farms. Tobacco and cotton fields and tea estates
were reportedly particularly under threat (Nyasaland Dept. Ag. An. Rept. 1924,
p. 5). In Nyasaland by the mid-1910s, both Europeans and Africans believed that
each others land use practices had caused an ecological crisis (Mlia 1987, p. 4),
including an increase in soil erosion.
During the 1920s regional awareness of soil erosion as a problem that could
be controlled and prevented resulted in a number of government and professional
publications, conferences and exchange among government officers in South Africa,
Southern Rhodesia (modern Zimbabwe) and Nyasaland (modern Malawi) (Report
of Select Committee 1914; Torrance 1919; Showers forthcoming). The release of
the South African Drought Investigation Commissions report in 1923 confirmed
that European land management practices were causing the destruction of the soil
and hydrological regimes, and that erosion was a national problem.
The 1920s also marked the beginning of systematic and sustained investiga-
tions of the southern African environment. Experiment stations were established
initially to identify the best crops, varieties, and management practices for European
settlers. Regional departments of agriculture became increasingly interested in iden-
tifying practices to limit surface wash (sheet erosion) because it was understood
to be linked to losses of soil fertility. Sheet not gully erosion was the major
concern in most places (Showers forthcoming). A government research programme
to characterise Nyasalands soils was implemented in the mid-1920s. Beginning
in 1924, soils were surveyed and samples were taken for chemical and physical
analysis. The results were used to construct an erosivity index (susceptibility of a
soil to erode). Both the physical and chemical properties of major Nyasaland soils
and the erosivity index were published in the 1930 Department of Agriculture
Annual Report (Nyasaland Dept. Ag. An. Rept. 1925, p. 26; 1930, p. 21). It was
158
Kate B. Showers
also noted that soils with a distinct textural change were most susceptible to ero-
sion. In the 1930s and 1940s increasing official concern about soil erosion was
reflected in an increase in government and professional publications; research to
understand erosion processes; and programmes for farmers and ranchers, in an
attempt to prevent erosion, or mitigate its consequences by restoring eroded land
(Showers 1989; Showers forthcoming).
Europeans settled as farmers and ranchers in modern South Africa, Swaziland,
Botswana (Bechuanaland), Zimbabwe (Southern Rhodesia) and Malawi (Nyasaland).
Most were literate, and had access to both regional publications and professional and
social contacts with government officials. However, these land users did not appear
to share official concern about soil erosions threat to the future of the landscape.
Although European farmers in Southern Rhodesia were visited by the Agricultural
Engineer to advise on soil conservation in the late 1920s, in 1938 only two districts
had over 40 per cent of the cultivated land protected by contour ridges (S. Rhodesia
An. Rept. 1929, p. 12; S. Rhodesia An. Rept. 1930, p. 39; S. Rhodesia An. Rept.
1931, p. 13; S. Rhodesia Dept. Ag. and Lands An. Rept. 1939, p. 28; S. Rhodesia
Dept. Ag. and Lands An. Rept. 1939, p. 3). The 1939 report of the Commission
to Enquire into the Preservation of the Natural Resources stated that soil erosion
was still a serious problem on European farmland (McIlwaine 1939). By the end
of 1940, less than three per cent of white farmers in South Africa had elected to
participate in government programmes to build conservation works on their land
because only a minority of farmers and a very small percentage of townspeople
were concerned about soil erosion (Chief, Div. Soil and Veld Cons.1941; Ross 1963,
p. 19). Despite the creation of a Natural Resources Board in Swaziland, meetings,
and advice to employ conservation techniques, the government had to resort to
fines to enforce reclamation and erosion control measures on European farms in
the mid-1950s (Swaziland An. Rept. 1953, p. 15; Swaziland An. Rept. 1955, p.
15; Swaziland An. Rept. 1958, p. 17).
African farmers were equally uninterested in the Departments of Agricultures
soil conservation techniques. Their lack of enthusiasm for the various methods was
noted in South Africa, Basutoland, Swaziland and Southern Rhodesia (Engineer
to Chief Native Commissioner 1941; Basutoland Dept. Ag. An. Rept. 1937/38,
p. 76; Basutoland Dept. Ag. An. Rept. 1938/39, p. 77; Basutoland Dept. Ag. An.
Rept. 1948, p. 31; Weinmann 1975, p. 206; Swaziland Col. An. Rept. 1950, p. 7;
Swaziland Col. An. Rept. 1951, p. 5; Swaziland Col. An. Rept. 1954, p. 4). However,
African land users did not have a choice about the adoption of these conservation
practices. Colonial officials decided what would be best for African land and for
Africans. Each colonial government devised a programme that it felt suited its
colonys unique conditions. Most included some kind of engineering structures,
like diversion ditches and terraces (which were referred to as contour banks, ridges
or bunds), grass strips and mechanisms for the regulation of grazing.
159
A History of African Soil
Africans have not been passive in the face of the social, economic and
environmental changes that occurred with the arrival of Europeans and their
technologies. They made changes in their land use practices. In many places, as
fallowing periods were constrained or eliminated, they were replaced by manuring.
During the 1990s this process was studied in the semi-arid region of West Africa
(including significant parts of Burkina Faso, Chad, Gambia, Mali, Mauritania,
Niger and Senegal), where soils are inherently low in plant nutrients, soil organic
matter and water retention capacity (Williams 1999, pp. 15, 17). Manure and
fallow, not fertiliser, were the major means of maintaining soil fertility. Manure
was applied either by hand-spreading on fields, or by corralling stock on a field at
night. Except on small fields, manure was placed on specific spots within a field
deemed to need nutrients rather than the whole field. Over time, each area of the
field received a manure application. The farmers belief that manure had long-term
effects and was not needed each year (Williams 1999) has been confirmed in stud-
ies such as by Powell et al. (1998). Research has shown that manure augments soil
organic matter contents, raises soil pH, improves cation exchange capacity (CEC)
and water-holding capacity of soils. When used in combination with inorganic
fertiliser, especially nitrogen, it reduces the fertilisers negative effects, particularly
acidification, and increases removal of nutrients other than those supplied by the
fertiliser (Williams 1999, p. 15). However, not every farmer was found to have
access to as much manure as s/he would like (Powell et al. 1998, p. 260; Williams
1999, p. 19).
When livestock are kept in stalls, as is advocated for modern production
systems, instead of corralled on a field as in traditional systems, urine is lost through
runoff from the barn and in seepage. The remaining manure must be handled,
stored and transported before spreading on fields. Researchers warn that the result
of such a switch in management practices could reduce the amounts and types of
nutrients available for recycling and increase nutrient losses, thus jeopardising the
long term productivity of the soil (Powell et al. 1998, p. 260)
b. Livestock Management
European methods of livestock management involved concentrated use of grazing
areas. Fences contained the animals, and there were limited locations for the stock
to drink. This approach produced overstocked pastures and heavy animal traffic
around corrals and watering holes all of which reportedly caused soil erosion
(Hailey 1938, p. 1080; Tempany 1949, p. 1; Jacks and Whyte 1938, p. 71; Ross
1963, p. 12). As Europeans claimed land for farms and ranches, first in southern
Africa and then in East Africa (Kenya in particular), African grazing areas were
reduced and pastures became overgrazed. Ward et al. (2000) addressed concerns
about land degradation on a communal ranching area in arid Namibia. Otjimb-
ingwes mean annual rainfall of 165.4 mm precludes crop production; livelihoods
161
A History of African Soil
could only come from livestock management. Using archival materials (including
photographs) and oral histories as well as soil and vegetational sampling, the re-
searchers concluded that it was rainfall, not stock numbers or management, that
determined the condition of the landscape. Even around watering points used
for 150 years, changes in soil quality (as measured by soil nitrogen, phosphorus,
organic carbon, and water-holding capacity by bioassay) could not be detected
(Ward et al. 2000, p. 351).
c. Architecture/Urbanisation
European urban areas in Africa were established for reasons of commerce, trans-
portation or administration, and also grew up around mining and industrial
enterprises. Some were expansions of existing African settlements, while others
were completely new. As with African settlements before them, European towns
affected soil by covering it, quarrying it for use in construction, and stimulating
the creation of roads and paths. Many European buildings were constructed with
brick or cement, which required local quarrying operations. Waste management
also affected soils.
Transportation to and from urban areas posed problems of compaction and
drainage. As unimproved roads became too rutted, wagons and later automobiles
and trucks created new tracks in the grass alongside the existing one. In this way,
small paths became large, rutted, eroding spaces (Kruger 1882 in Germond 1967,
p. 407; Assistant Engineer, 1934). Faulty road and railway drainage systems also
caused gully erosion in South Africa, Basutoland and Southern Rhodesia (Watt
1913; Stewart 1917; S. A. Dept. Ag. and For. 1941, p. 18; Hailey 1938, p. 1058;
Jacks and Whyte 1938, p. 70; Pentz 1940, p. 4; Proc. Conf. Soil Erosion 1945;
Showers 1989, pp. 270, 275; Phiri 2003, p. 12; Showers 2005, pp. 143, 1501) and
railway constructions demand for wooden sleepers encouraged deforestation and
subsequent erosion, especially in savanna regions that did not have many trees.
Municipal waste disposal domestic and industrial was generally un-
derstood to be a process of removal. Treatment was rarely a consideration. Sewage
and industrial effluents were dumped in streams and the ocean; abandoned quar-
ries were used as disposal sites, and, in some semi-arid regions, liquid waste was
simply dumped on the ground. Most urban areas were built with septic, rather
than centralised water borne, sewage systems. The septic systems overwhelmed the
soils, resulting in subsoil pollution (for discussion and tables of African urban and
industrial waste disposal and pollution see Showers 2002).
Phiri (2003) assessed the environmental impact of the establishment and
growth of the Southern Rhodesian town of Umtali (modern Mutare, Zimbabwe).
Established as a base from which gold mining operations could be expanded,
Umtali displaced an indigenous African population to other, less productive land,
which ultimately became overused. As mines were dug, soil was disturbed directly
162
Kate B. Showers
and indirectly. To prospect for gold, miners burned the grass for easier access to
the earth, and no attention was paid to maintaining soil cover against erosion. The
combined mining and urbanisation processes led to deforestation: trees were cut for
timbers in the mines, sleepers for the expanding railway track, and for construction.
Erosion resulted (Phiri 2003, p. 6).
As mining operations encroached on the original site of Old Umtali, the
town was moved and expanded to New Umtali. Soils were again disturbed as
buildings were built, and roads laid. A new town with new and better buildings
required more wood and of higher grade. Brick was a desired building material, so
large scale soil excavation began for brick making. The new towns roads were not
well constructed, and drained badly. Domestic, milling and mining wastes posed
disposal problems that were never properly resolved. Soil disruption and erosion,
waste accumulation, and pollution accompanied Umtalis development. As modern
Mutare, it faced the same problems in 1995 although on a different scale that
Old and New Umtali had in the late nineteenth and early twentieth centuries.
d. Mining
It was the 1867 discovery of diamonds in northern Cape Colony, South Africa
that triggered the mineral revolution, despite coal having been found in Natal
Colony in the 1840s (Wilson and Thompson 1971, p. 11). Large-scale European
mining in Africa began with diamonds, but it was institutionalised as a compo-
nent of economic development when the Witwatersrand was proclaimed a gold
mining area in 1886 (Wilson and Thompson 1971, p. 13). Although Southern
Africa has Africas greatest concentration of non-petroleum minerals, by the end
of the twentieth century every African country had large- and small-scale mines
(Griffiths 1984, p. 126). In the 1920s and 1930s mining operations began around
the continent, including small-scale gold mines in Cameroon, mines for iron and
diamonds in Sierra Leone, diamonds on the west coast of southern Africa, and
copper in Zambia (Desmet and Cowling 1999, p. 35; Grove, A.T. 1970, p. 109;
Mobbs 1998; Lewis and Berry 1988, p. 369). A further wave of mine develop-
ment occurred in the 1960s and 1970s, as African nations gained independence.
Mining activities disturb surface soils, and create sources of pollution, dispersed
by water and wind to surrounding soils undisturbed by actual mining operations
(Lewis and Berry 1988, p. 367).
When minerals are quarried, the unwanted material surrounding the miner-
als (tailings) is discarded in piles. This can involve a considerable amount of soil;
northern Botswanas rich Orapa Mine produces 0.89 carats of diamonds for every
ton of soil excavated (Lewis and Berry 1988, p. 367). In South Africa, until the
1920s, tailings from the increasingly deep and urban Witwatersrand gold mines were
deposited in dry form on so-called sand dumps; after 1921 the deposits were slurries
on slimes dams (Groves 1974, p. 296). By the mid 1970s the Witwatersrand alone
163
A History of African Soil
had 1,200 ha of sand dumps and 6,800 ha of slimes dams (Groves 1974, p. 296).
Until stabilisation measures began to be taken in the mid-1960s, dust blew at such a
rate from these dams and dumps that visibility was sometimes reduced to 10 metres
(Groves 1974, p. 296). The dust settled on soils in surrounding areas. Tailings piles
also produced sulphuric acid, which travelled in rainwater runoff to surrounding
soil and water courses. On South Africas arid southwest coast, opencast mining for
diamonds since 1929 has resulted in large overburden dumps and other disturbed
areas (Desmet and Cowling 1999, p. 35). In contrast to the acidic (low pH) tailings
from gold mines, these overburden dumps are composed of dune sands and marine
deposits with high pHs and high salt contents (Desmet and Cowling 1999, p. 35).
Left on their own, the slimes dams, sand dumps and overburden dumps were not
completely revegetated. In the mid-1960s attention turned to stabilising the urban
tailings piles, but only recently has attention turned to the rural arid overburden
dumps, some of which have remained bare since the 1970s (Desmet and Cowling
1999, p. 36). By 1990 South Africa had 450 mine dumps (Koch et al. 1990, p.
49). For photographs of mines in the Orange River catchment, including coastal
diamond mines, see the South African Department of Water Affairs Orange River
website at http: //www.dwaf.gov.za/orange/Low_Orange.htm.
In Northern Rhodesia (modern Zambia), copper mining began in 1921.
From 1921 to 1957, when hydroelectricity became available, the surrounding
forest was cleared to fuel the thermal power stations supplying the mines. Some
118,857 hectares were cleared between 1947 and 1964. Not only did this disrupt
the local shifting cultivation system by removing land for fallow, but it also changed
the water balance of the soil (Lewis and Berry 1988, pp. 370, 371). The mines
themselves produced huge open pits and tailings piles, and the refining process
released copper and lead into the air, which settled on the surrounding landscape
(Lewis and Berry 1988, p. 371).
Southern Africas mining industry is, perhaps, the biggest and best docu-
mented, but the processes are the same around the continent. Mining operations
make holes and generate waste, they require large amounts of electricity and water
for operation, and they require a population of workers. Mines are, as Phiri (2003)
pointed out, an agent of urbanisation, and create all of the environmental effects
of any urban centre.
e. Trade
African local, regional and intercontinental importing and exporting has an ancient
history. Ceramics, metal ore and finished metal products were widely circulated.
Livestock played a significant role in transport and as objects of trade. Shaw (1995)
discussed camels in Roman North Africa and the Sahara, and Kreike (2004) out-
lined the extensive local and long-distance cattle trade in the late-eighteenth and
nineteenth centuries carried out by the Ovakwanyama, residents of the Ovambo
164
Kate B. Showers
flood plain (modern northern Namibia and southern Angola). In contrast to pre-
colonial economies in Africa, those constructed by Europeans placed increasingly
heavy pressures on soils. With the sale of agricultural and livestock products, nutrient
cycling systems were broken, and replacements were not always made. As indus-
trial processes increased the demand for both the amount and kind of minerals,
prospecting and mining expanded, magnifying the effects on soil discussed above.
Urban and industrial areas required power, which resulted in rich alluvial soils
being submerged by hydroelectric dams (Showers 2001). Finally, increased trade
required increased infrastructure from transportation to urbanisation with its
attendant soil effects.
of burning is the return of nutrients from vegetation to soil in the form of ash. As
mentioned above, fires that burned quickly did not affect soil below the top 5cm.
under either forest or savanna. Soil properties did change, however, when wood was
deliberately piled to create an intense heat, as in the chitemene system. The cation
exchange capacity was increased, and could persist for as long as 16 years. Burning
in this instance could be considered to be a temporary transforming process for
the increased CEC would not persist long after cessation of the burning sequence.
From a soil chemical perspective, burning produced ash that released cations and
anions into the soil solution. This affected the acidity of soils (making them less
acid), and the nutrient availability to plants. The pH change, which affects soil mi-
crobial communities as well as plant roots, is temporary. However, the consequence
of burning that receives the most attention is the loss from the soil-plant system
of carbon, nitrogen and sulphur. If these losses can be compensated for over time
by plants, micro-organisms, or rainfall and lightning (nitrogen), then burning can
be seen as another aspect of nutrient cycling, without net loss.
There is no question that with the arrival of Europeans and their land use
practices and technologies, soils changed dramatically in some places. European
settlers and plantation owners displaced Africans from land, thus limiting previ-
ously unbounded extensive land use systems and increasing human and livestock
populations on the remaining land. This concentration had implications for the
stability of fallow systems and the stimulation of soil erosion, as discussed earlier.
The newly arrived technologies were untested on African soils (although arguments
have been made for the pre-European rejection of ploughs in favour of less soil-dis-
turbing digging sticks by Africans living on structurally fragile soils of West Africa).
Whether employed by Europeans or Africans, the new agricultural technologies
brought change. Certainly the soil conservation ideas imported from the United
States in the early twentieth century proved to be highly destructive to many of the
soils they were meant to protect. (See Showers 2005 and forthcoming for discussion
of soil conservation technology transfer from USA to southern Africa.)
Each landscape was affected differently, but in places, soil textures, structures,
chemistry and hydrology were changed some irreversibly. As population densities
increased and urban areas were created or expanded, soil cover was altered; ploughs,
monocrops and continuous cultivation lowered soil organic matter levels and nutri-
ent supplies; and agricultural chemicals (synthetic fertilisers, pesticides, herbicides)
changed soil chemistries and microbial populations temporarily or permanently.
Soil hydrologies were changed by mining operations, compaction and cover associ-
ated with urban areas and roads, and by gullies. Soils lost surfaces through erosion.
Mines and industries released toxins to the soil environment, some of which persist
for years, as do many agricultural chemicals. Soil was the object of concern only
when it presented a problem, such as when it eroded, produced lower than expected
yields, or failed to accept all of the waste products deposited on or into it. Soil
166
Kate B. Showers
conservation was seen as a technology that could remedy land shortage and allow
the persistence of certain European practices, rather than an expression of concern
about soil bodies themselves. It is obvious that soil properties, and in some places
entire soil bodies, were changed substantially in the twentieth century.
A true history of African soils will have to be many histories, with each
regions soils considered. Documentation will be needed to show the mechanism(s)
by which a particular soil body has been changed. Analysis must consider whether
the changes identified were permanent or temporary in nature, as well as whether
they are reversible or irreversible. A tool for distancing researchers from the logics
of dominant narratives (crisis or otherwise) is historical environmental impact as-
sessment (HEIA), which focuses attention on landscape dynamics and facilitates the
use of data from a wide variety of sources (Showers 1996). Given the tremendous
paucity of information about most of the African continent, local soil knowledge
systems ought to be major sources. Even when there are simply no data, and no
sources can be imagined, the temptation to extrapolate or generalise must be resisted.
Clear statements of gaps are more useful than projections and assumptions. No
information is a kind of data, sometimes the only legitimate kind.
The twenty-first century dawns on an Africa with different soil conditions
than at the start of the twentieth century, and knowledge of Africas soils remains
incomplete including its history in detail. Proper histories of African soils could
make a useful contribution to addressing the very real soil concerns of the twenty-
first century.
References
Abrahams, P.W. 1997. Geophagy (Soil Consumption) and Iron Supplementation in Uganda.
Tropical Medicine and International Health, 2(7): 61723.
Adams, W.M. 1992. Wasting the Rain: Rivers, People and Planning in Africa. Minneapolis:
University of Minnesota Press.
Alemayehu, Million. 1996. Traditional Ditches in Northern Showa, the Ethiopian High-
lands. In Chris Reij, Ian Scoones and Camilla Toulmin, Sustaining the Soil: Indigenous
Soil and Water Conservation in Africa (London: Earthscan), pp. 1639.
Amborn, Hermann. 1989. Agricultural Intensification in the Burji-Konso Cluster of South-
Western Ethiopia. Azania, 24: 7183.
Asrat, Kebede, Kederalah Idris and Mesfin Semegen. 1996. The Flexibility of Indigenous
Soil and Water Conservation Techniques: A Case Study of the Harerge Highlands,
Ethiopia. In Chris Reij, Ian Scoones and Camilla Toulmin, Sustaining the Soil: Indigenous
Soil and Water Conservation in Africa (London: Earthscan), pp. 15662.
Assistant Engineer. 1934. Answers to Question 16, A.W. Pim Commission of Inquiry,
1944. Assistant Engineer, Department of Public Works. Public Records Office DO
119/1051.
167
A History of African Soil
Chief, Division of Soil and Veld Conservation. 1941. To Secretary for Agriculture and Forestry,
No.R.1435/91, September 1941. South African Archives LDB R4250 Vol. 1.
Childs, S. Terry and Eugenia W. Herbert. 2005. Metallurgy and its Consequences. In Ann
Brower Stahl (ed.), African Archaeology: A Critical Introduction (Malden, MA: Blackwell
Publishing), pp. 276300.
Coupland, R.T. 1979. Natural Temperate Grasslands: Use and Management. In R.T.
Coupland (ed.), Grassland Ecosystems of the World (Cambridge: Cambridge University
Press), pp. 10711.
Davis, Diana K. 2005. Potential Forests: Degradation Narratives, Science and Environmental
Policy in Protectorate Morocco, 19121956. Environmental History, 10: 21138.
Denyer, Susan. 1978. African Traditional Architecture. New York: Africana Publishing
Company.
Desmet, P.G. and R.M. Cowling. 1999. Patch Creation by Fossorial Rodents: A Key Proc-
ess in the Revegetation of Phytotoxic Arid Soils. Journal of Arid Environments, 43(1):
3545.
Dokuchaev Central Soil Museum, 20002003. V.V. Dokuchaev (18461903). http: //
www.soilmuseum.by.ru/english/dokuchaev_eng.htm.
Dreschel, P., D. Kunze and F. Penning de Vries. 2001. Soil Nutrient Depletion and Popula-
tion Growth in Sub-Saharan Africa: A Malthusian Nexus? Population and Environment,
22(4): 411423.
Eggert, Manfred K.K. 2005. The Bantu Problem and African Archaeology. In Ann Brower
Stahl (ed.), African Archaeology: A Critical Introduction (Malden, MA: Blackwell Pub-
lishing), pp. 30126.
Encyclopaedia Britannica. 1929. Laterite. Vol. 13: 740.
Endfield, Georgina H. and David I. Nash. 2002. Drought, Desiccation and Discourse: Mis-
sionary Correspondence and Nineteenth-Century Climate Change in Central Southern
Africa. The Geographical Journal, 168(1): 3347.
Engineer to Chief Native Commissioner. 5 November 1941. No. 115/8. South African
Archives, NTS 15/276 Vol. II.
Eswaran, Hari. 1988. Taxonomy and Management Related Properties of the Red Soils
of Africa. In The Red Soils of East and Southern Africa, Proceedings of an International
Symposium, Harare, Zimbabwe, 2427 February 1986, Rapport Manuscript. IDRC-
MR170e (Ottawa: International Development Research Centre), pp. 123.
Eswaran, H., F.H. Beinroth, J. Kimble. and T. Cook. 1992. Soil Diversity in the Tropics:
Implications for Agricultural Development. In R. Lal and P.A. Sanchez, Myths and
Science of Soils of the Tropics, Soil Science Society of America Special Publication No.29
(Madison: Soil Science Society of America), pp.116.
Faniran, A. and O. Areola. 1978. Essentials of Soil Study: With Special Reference to Tropical
Areas. London: Heinemann.
FAO/Unesco.1977. Soil Map of the World, 1: 5 000 000, Vol. 6, Africa. Paris: Unesco.
Germond, R.C. 1967. Chronicles of Basutoland. Morija: Morija Sesuto Book Depot.
169
A History of African Soil
Gifford-Gonzalez, Diane. 2005. Pastoralism and its Consequences. In Ann Brower Stahl
(ed.), African Archaeology: A Critical Introduction (Malden, MA: Blackwell Publishing),
pp. 187224.
Giles-Vernick, Tamara. 2002. Cutting the Vines of the Past: Environmental Histories of the
Central African Rain Forest. Charlottesville: University Press of Virginia.
Goldammer, J.G. and C. de Ronde (eds). 2004. Wildland Fire Management Handbook for
Sub-Sahara Africa. Frieburg: Global Fire Management Centre/Oneworld Books.
Gradwell, M.W. 1968. Compaction of Pasture Topsoils under Winter Grazing. In Transac-
tions 9th International Congress of Soil Science (Adelaide Australia, 1968) (New York City:
American Elsevier Publishing Co.), vol.3, pp. 42935.
Griffiths, Ieuan Ll. 1984. An Atlas of African Affairs. New York: Methuen.
Grove, A.T. 2000. The African Environment, Understood and Misunderstood. In Douglas
Rimmer and Anthony Kirk-Greene (eds), The British Intellectual Engagement with Africa
in the Twentieth Century (New York: St. Martins Press), pp. 179206.
Grove, A.T. 1970. Africa South of the Sahara, 2nd edition. London: Oxford University
Press.
Grove, A.T. and J.E.G. Sutton. 1989. Agricultural Terracing South of the Sahara. Azania,
24: 11322.
Grove, Richard. 1989. Scottish Missionaries, Evangelical Discourses and the Origins of
Conservation Thinking in Southern Africa 18201900. Journal of Southern African
Studies, 15(2): 16387.
Groves, J.E. 1974. Reclamation of Mining Degraded Land. South African Journal of Sci-
ence, 70 (10): 2969.
Guyer, Jane and Samuel M.E. Belinga. 1995. Wealth in People as Wealth in Knowledge:
Accumulation and Composition in Equatorial Africa. Journal of African History, 36(1):
91120.
Hailey, Lord. 1938. An African Survey: A Study of Problems Arising in Africa South of the
Sahara. London: Oxford University Press.
Hamza, M. Ait. 1996. The Mastery of Water: SWC Practices in the Atlas Mountains of
Morocco. In Chris Reij, Ian Scoones and Camilla Toulmin (eds) Sustaining the Soil:
Indigenous Soil and Water Conservation in Africa (London: Earthscan), pp. 447.
Haynes, R.J. and G.S. Francis. 1993. Changes in Microbial Biomass C, Soil Carbohydrate
Composition and Aggregate Stability Induced by Growth of Selected Crop and Forage
Species Under Field Conditions. Journal of Soil Science, 44 (4): 66575.
Heusch, B. 1981. Sociological Constraints in Soil Conservation: A Case Study, the Rif
Mountains, Morocco. In R.P.C. Morgan (ed.), Soil Conservation: Problems and Prospects
(London: John Wiley and Sons), pp. 41924.
Hilson, Gavin. 2002. Small-Scale Mining in Africa: Tackling Pressing Environmental Problems
with Improved Strategy. Journal of Environment and Development, 11(2): 14974.
Hiol Hiol, Franois; Dieudonn Ndoum Mbeyo and Franois Tchala Abina. 1996. Traditional
Soil and Water Conservation Techniques in the Mandara Mountains, Northern Cam-
eroon. In Chris Reij, Ian Scoones and Camilla Toulmin, Sustaining the Soil: Indigenous
Soil and Water Conservation in Africa (London: Earthscan), pp. 191201.
170
Kate B. Showers
Hoffman, Irene, Dirk Gerling, Usman B.Kyiogwom, and Anke Man-Bielfeldt. 2001. Farm-
ers Management Strategies to Maintain Soil Fertility in a Remote Area in Northwest
Nigeria. Agriculture, Ecosystems and Environment, 86: 26375.
Hopkins, A.G. 1973. An Economic History of West Africa. London: Longman.
Igbokwe, E.M. 1996. A Soil and Water Conservation System Under Threat: A Visit to Maku,
Nigeria. In Chris Reij, Ian Scoones and Camilla Toulmin, Sustaining the Soil: Indigenous
Soil and Water Conservation in Africa (London: Earthscan), pp. 21927.
Ikpe, F.N., J.M. Powell, N.O. Isirimah, T.A.T. Wahua and E.M. Ngodigha. 1999. Effects
of Primary Tillage and Soil Amendment Practices on Pearl Millet Yield and Nutrient
Uptake in the Sahel of West Africa. Experimental Agriculture, 35(4): 43748.
Jacks, G.V. and R.O. Whyte. 1938. Erosion and Soil Conservation, Herbage Publication Series
Bulletin No. 25. Aberystwyth: Imperial Bureau of Pastures and Forage Crops.
Jacobs, Nancy. 2003. Environment, Power and Injustice: A South African History. Cambridge:
Cambridge University Press.
Johns, T. and M. Duquette. 1991. Detoxification and Mineral Supplementation as Func-
tions of Geophagy. American Journal of Clinical Nutrition, 53(2): 44856.
Johnson, Douglas H. and David M. Anderson.1988. The Ecology of Survival: Case Studies
from Northeast African History. Boulder, CO: Westview Press.
Kass, Donald C., H. David Thurston, and Kenneth Schlather. 1999. Sustainable Mulch-
Based Cropping Systems with Trees. In Louise E. Buck, James P. Lassoie and Erick
C.M. Fernandes (eds), Agroforestry in Sustainable Agricultural Systems (Boca Raton, FL:
CRC Press), pp. 36179.
Killingray, David. 2000. Colonial Studies. In Douglas Rimmer and Anthony Kirk-Greene
(eds), The British Intellectual Engagement with Africa in the 20th Century (New York:
St. Martins Press).
Kjekshus, Helge. 1996. Ecology Control and Economic Development in East African History:
The Case of Tanganyika 18501950, 2nd impression with new introduction. London:
James Currey.
Klieman, Kairn A. 2003. The Pygmies were our Compass. Bantu and Batwa in the History of
West Central Africa, Early Times to c. 1900 C.E. Portsmouth, N.H.: Heinemann.
Koch, Eddie, Dave Cooper, and Henk Coetzee. 1990. Water, Waste and Wildlife: The Politics
of Ecology in South Africa, Penguin Forum Series. London: Penguin Books.
Krasilnikov, Pavel V. and Joe A. Tabor. 2003. Perspectives on Utilitarian Ethnopedology.
Geoderma, 111: 197215.
Kreike, Emmanuel. 2004. Re-creating Eden: Land Use, Environment and Society in Southern
Angola and Northern Namibia. Portsmouth, NH: Heinemann.
Krenzer, E.G., Jr., Chaw Foh Chee, and J.F. Stone. 1989. Effects of Animal Traffic on Soil
Compaction in Wheat Pastures. Journal of Production Agriculture, 2(3): 2469.
Lal, R. and P.A. Sanchez. 1992. Myths and Science of Soils of the Tropics, Soil Science Society
of America Special publication No. 29. Madison: Soil Science Society of America.
171
A History of African Soil
Moore, Henrietta L. and Megan Vaughn. 1994. Cutting down Trees: Gender, Nutrition and
Agricultural Change in the Northern Province of Zambia, 18901900. Portsmouth, NH:
Heinemann.
Mpofu, T.P.Z. 1987. History of Soil Conservation in Zimbabwe. In History of Soil Con-
servation in the SADCC Region, Report No. 8. Co-ordination Unit. Maseru: SADCC
Soil and Water Conservation and Land Utilization Programme.
Nachtergaele, F.O. undated. From the Soil Map of the World to the Digital Global Soil and Ter-
rain Database, 19602002. http://www.itc.nl/rossiter/Docs/WRB/SoilMapWold.pdf
National Geographic Society. 1990. National Geographic Atlas of the World. Washington,
D.C.: National Geographic Society.
Neumann, Katharina. 2005. The Romance of Farming: Plant Cultivation and Domestica-
tion in Africa. In Ann Brower Stahl (ed.), African Archaeology: A Critical Introduction
(Malden, MA: Blackwell Publishing), pp. 24975.
Niemeijer, David and Valentina Mazzucato. 2003. Moving beyond Indigenous Soil Taxono-
mies: Local Theories of Soils for Sustainable Development. Geoderma, 111: 40324.
Nyamapfene, K.W. 1987. A Short History and Annotated Bibliography on Soil and Water
Conservation, Report No. 12. Maseru: Coordination Unit, SADCC Soil and Water
Conservation and Land Utilization Programme.
Nyasaland. Department of Agriculture Annual Report for the Year: 1924, 1925, 1930. Zomba:
Government Printer.
Nye, P.H. and D.J. Greenland. 1960. The Soil under Shifting Cultivation, Technical Com-
munication No. 51. Harpenden: Commonwealth Bureau of Soils.
Oliver, Roland. 1991. The African Experience: Major Themes in African History from Earliest
Times to the Present. New York: IconEditions.
Osunade, M. and C. Reij. 1996. Back to the Grass Strips: A History of Soil Conservation
Policies in Swaziland. In C. Reij, I. Scoones and C. Toulmin (eds), Sustaining the Soil:
Indigenous Soil and Water Conservation in Africa (London: Earthscan).
Payton, R.W., J.F. Barr, A. Martin, P. Sillitoe, J.F. Deckers, J.W. Gowing, N. Hatibu, S.B.
Naseem, M. Tenywa and M.I. Zuberi. 2003. Contrasting Approaches to Integrating
Indigenous Knowledge about Soils and Scientific Soil Survey in East Africa and Bang-
ladesh. Geoderma, 111 (34): 355.
Pendleton, Don. 1982. Range Use and Management of Great Plains Soils. In Andrew
R. Aandahl (ed.), Soils of the Great Plains (Lincoln: University of Nebraska Press), pp.
1227.
Pentz, J.A. 1940. Soil Erosion Survey of the Reclamation Area in Natal, Veld Management
and Pasture Research Series Memoir No. 2, Science Bulletin No. 212, Union of South
Africa Department of Agriculture and Forestry. Pretoria: Government Printer.
Phiri, Christopher. 2003. Environmental Aspects of Umtalis Urbanisation: 18901923: An
Analysis. Paper presented at the International Conference on Forest and Environmental
History of the British Empire and Commonwealth, Centre for World Environmental
History, University of Sussex, March 1921, 2003. Forthcoming in Richard Grove and
Vinita Damodaran (eds), Empire and Environment: The British Empire and the Natural
World, 1600 2000 (India: Oxford University Press).
173
A History of African Soil
Platzky, Laurine and Cheryl Walker. 1985. The Surplus People: Forced Removals in South
Africa. Johannesburg: Ravan Press.
Powell, J.M., S. Fernndez-Rivera, P. Hiernaux, and M.D. Turner. 1996. Nutrient Cycling
in Integrated Rangeland/Cropland Systems of the Sahel. Agricultural Systems, 52(2/3):
14370.
Powell, J.M., F.N. Ikpe, Z.C. Somida and S. Fernndez-Rivera. 1998. Urine Effects on
Soil Chemical Properties and the Impact of Urine and Dung on Pearl Millet Yield.
Experimental Agriculture, 34(3): 25976.
Proceedings of a Conference on Soil Erosion caused by Roads and Railways, October 1945.
Pretoria. South African Archives LDB R 4250/1.
Report of the Select Committee on Droughts, Rainfall and Soil Erosion together with
proceedings of the committee, minutes of evidence and appendix. 1914. South African
Parliament, Senate, 4th Session, 19 June 1914. Cape Town: Government Printer.
Richards, Paul. 1985. Indigenous Agricultural Revolution: Ecology and Food Production in
West Africa. London: Hutchinson.
Roe, Emery. 1991. Development Narratives, or Making the Best of Blueprint Development.
World Development, 19(4): 287300.
Roe, Emery. 1995. Except-Africa: Postscript to a Special Section on Development Narra-
tives. World Development, 23(6): 10659.
Roland, J.W. 1938. Pasture Research in South Africa, Department of Agriculture and Forestry,
Progress Report No. 1. Pretoria: Government Printer.
Ross, J.C. 1963. Soil conservation in South Africa: A Review of the Problem and Developments
to Date, Department of Agriculture Technical Services. Pretoria: Government Printer.
Russell, E.W. 1973. Soil Condition and Plant Growth, 10th edition. London: Longman.
Sayer, Jeffrey A., Caroline S. Harcourt and N. Mark Collins (eds), 1992. The Conservation
of Tropical Forests: Africa. New York: Simon and Schuster.
Schantz, H.L. and C.F. Marbut. 1923. Vegetation and Soils of Africa, American Geographical
Society Research Series No. 13, National Research Council and American Geographical
Society. New York: AMS Press (Reprinted 1971).
Scoones, Ian. 1997. The Dynamics of Soil Fertility Change: Historical Perspectives on
Environmental Transformation from Zimbabwe. The Geographical Journal, 163(2):
1619.
Shaw, Brent D. 1995. The Camel in Roman North Africa and the Sahara: History, Biology
and Human Economy. In Brent D. Shaw, Environment and Society in Roman North
Africa (Aldershot: Variorum), pp. 663721.
Shahack-Gross, Ruth, Fiona Marshall and Steve Weiner. 2003. Geo-Ethnoarchaeology of
Pastoral Sites: The Identification of Livestock Enclosures in Abandoned Maasai Settle-
ments. Journal of Archaeological Science, 30(4): 43959.
Showers, Kate B. 1989. Soil Erosion in the Kingdom of Lesotho: Origins and Colonial
Response, 1830s1950s, Journal of Southern African Studies 15(2): 26386.
174
Kate B. Showers
Showers, Kate B. 1994. Early Experiences of Soil Conservation in Southern Africa: Segre-
gated Programs and Rural Resistance. Working Paper No. 184, African Studies Center.
Boston: Boston University.
Showers, Kate B. 1996. Soil Erosion in the Kingdom of Lesotho and Development of
Historical Environmental Impact Assessment. Ecological Applications, 6(2): 65364.
Showers, Kate B. 2001. Dams and Power in Southern Africa. In C.M. Miller, M. Cioc and
K.B. Showers (eds), Water and the Environment Since 1945: A Cross-Cultural Perspective,
History in Dispute Series, vol. 7 (Detroit: St. James Press), pp. 23641.
Showers, Kate B. 2002. Water Scarcity and Urban Africa: An Overview of Urban-Rural
Water Linkages. World Development, 30(4): 62148.
Showers, Kate B. 2003. Origins of Soil Conservation in British Africa, Pt. I: Southern
Africa through the 1920s. Paper presented at the International Conference on Forest
and Environmental History of the British Empire and Commonwealth, Centre for
World Environmental History, University of Sussex, March 1921, 2003. Forthcoming
in Richard Grove and Vinita Damodaran (eds), Empire and Environment: The British
Empire and the Natural World, 1600 2000 (India: Oxford University Press).
Showers, Kate B. 2005. Imperial Gullies: A History of Soil Erosion and Conservation in the
Kingdom of Lesotho. Athens: Ohio University Press.
Sillitoe, P. 1998. Knowing the Land: Soil and Land Resource Evaluation and Indigenous
Knowledge. Soil Use and Management 14: 18893.
Soil Survey Staff. 1951. Soil Survey Manual, USDA Handbook No. 18, Bureau of Plant
Industry, Soils and Agricultural Engineering, USDA. Washington DC: Government
Printing Office.
South African Department of Agriculture and Forestry. 1941. Report of the Departmental
Soil Erosion Committee. South African Archives LDB R4250 Vol. 1.
Southern Rhodesia. Annual Report for the year: 1929, 1930, 1931. Salisbury: Government
Printer.
Southern Rhodesia Department of Agriculture and Lands Report. 1939. Salisbury: Govern-
ment Printer.
Stahl, Ann Brower. 2005a. Introduction: Changing Perspectives on Africas past. In Ann
Brower Stahl (ed.) African Archaeology: A Critical Introduction (Malden, MA: Blackwell
Publishing), pp. 123.
Stahl, Ann Brower (ed). 2005b. African Archaeology: A Critical Introduction. Malden, MA:
Blackwell Publishing
Stromgaard, Peter. 1986. Early Secondary Succession on Abandoned Shifting Cultivators
Plots in the Miombo of South Central Africa. Biotropica, 18(2): 97106.
Stewart, G.A. 1917. Roads of Basutoland. Report, 4 May 1917. In Colonial Office: High
Commission for South Africa, Original Correspondence. Despatches: Basutoland 1917.
Public Records Office CO 417/593.
Sutton, J.E.G. 1984. Irrigation and Soil-Conservation in African Agricultural History with
a Reconsideration of the Inyanga Terracing (Zimbabwe) and Engaruka Irrigation Works
(Tanzania). The Journal of African History, 25(1): 2541.
Sutton, J.E.G. 1989. Towards a History of Cultivating Fields. Azania, 24: 98112.
175
A History of African Soil
Swaziland. Colonial Annual Report for the Year: 1946; 1950; 1951; 1952; 1953; 1954; 1955;
1956; 1958.
Tanner, C.B. and C.P. Mamaril. 1959. Pasture Soil Compaction by Animal Traffic. Agronomy
Journal, 51(6): 32931.
Temato, P. and K.R. Olson. 1997. Impacts of Industrialized Agriculture on Land in Bafou,
Cameroon. Journal of Soil and Water Conservation, 52(6): 4045.
Tempany, Harold A. 1949. The Practice of Soil Conservation in the British Colonial Empire,
Technical Communication No. 45. Harpendon: Commonwealth Bureau of Soil Sci-
ence.
Temu, A.E.M. and S. Bisanda. 1996. Pit Cultivation in the Matengo Highlands of Tanzania.
in Reij, Chris, Ian Scoones and Camilla Toulmin (eds) Sustaining the soil: Indigenous soil
and water conservation in Africa (London: Earthscan), pp. 14550.
Thompson, Robert Farris and Joseph Cornet. 1981. The Four Moments of the Sun: Kongo
Art in Two Worlds. Washington, DC: National Gallery of Art.
Tlou, Thomas and Alec Campbell. 1984. History of Botswana. Gaberone: Macmillan Bot-
swana Publishing Co.
Torrance, W. 1919. Observation on Soil Erosion, Union of South Africa Department of
Agriculture Bulletin 4. Pretoria: Government Printer.
Trore, P.A. 1994. Constraints on Small-Scale Mining in Africa. Natural Resources Forum,
18(3): 20712.
Trustees of Boston University. n.d. How Big is Africa? Outreach program. Boston: Boston
University African Studies Center.
Twerdoff, D.A., D.S. Chanasyk, E. Mapfumo, M.A. Naeth, and V.S. Baron. 1999. Impacts
of Forage Grazing and Cultivation on Near-Surface Relative Compaction. Canadian
Journal of Soil Science, 79(3): 46571.
Van der Vijver, C.A.D.M., P. Poot, and H.H.T. Prins. 1999. Causes of Increased Nutri-
ent Concentrations in Post-Fire Regrowth in an East African Savanna. Plant and Soil
214(12): 17385.
Vansina. Jan. 1990. Paths in the Rainforests: Toward a History of Political Tradition in Equato-
rial Africa. London: James Currey.
Ward, David, Ben T. Ngairorue, Andr Apollus and Hermanus Tjiveze. 2000. Perceptions
and Realities of Land Degradation in Arid Otjimbingwe, Namibia. Journal of Arid
Environments, 45: 33756.
Wardell, D. Andrew, Thomas Theis Nielsen, Kjeld Rasmussen and Cheikh Mbow. 2004.
Fire History, Fire Regimes and Fire Management in West Africa: an Overview. In J.G.
Goldammer, and C. de Ronde (eds), Wildland Fire Management Handbook for Sub-Sahara
Africa (Frieburg: Oneworld Books/Global Fire Management Centre), pp. 35081.
Warren, S.D., T.L. Thurow, W.H. Blackburn and N.E. Garza. 1986. The Influence of Live-
stock Trampling under Intensive Rotation Grazing on Soil Hydrologic Characteristics.
Journal of Range Management, 39(6): 4915.
Watt, W. Martin. 1913. The Dangers and Prevention of Soil Erosion. Rhodesia Agriculture
Journal X: 5.
176
Kate B. Showers
Verena Winiwarter
Villicandi enim scientia adeo super omnes pollebat, quae videlicet in aedificiis con-
struendis, iumentis et pecoribus educandis, agris serendis aliarumve rerum rusticarum
constare videtur quacunque cultura, quam tamen non usu constat eum didicisse, sed
arte, ut pene haec eo nemo curaret solertius nec his felicius abundaret.
[With reference to knowledge in the field of agriculture, this means e.g. in construc-
tion of buildings, in breeding cattle and smaller animals, in treating fields or in any
work of other rural matters, he was far ahead of all others. Almost no-one had more
skills and better success than him. Yet his knowledge did not stem from tradition,
but from studies, and thus almost nobody worked more circumspectedly and had
better success than him.] (Chapter 8 from the Life of Bishop Benno II of Osnabrueck,
written 10901100, my translation of the German translation in Kallfelz 1973).1
Bishop Benno II was a learned and well-travelled man, although or even because
he was no member of the aristocracy. In the eleventh century, ministeriales like
him could climb up the social ladder considerably. Benno was born in 1020 in
Swabia, and studied in Strasbourg and at the Reichenau, the famous monastery
1
Thanks to Christoph Sonnlechner for alerting me to this source.
179
Prolegomena to a History of Soil Knowledge in Europe
situated in Lake Constance. He made a pilgrimage to the Holy Land, was a member
of the cathedral school of Speyer, then became governor of the cathedral school
of Hildesheim, after which he was appointed chief administrator (vicedominus) of
the Goslar royal estate. After a short interlude in Cologne he became Bishop of
Osnabrueck by appointment of King Henry IV. The bishop worked in a diplomatic
mission during the uprising of the Saxons 107375. He founded the Benedictine
monastery of Iburg, where he spent the last years of his life, and died in 1088. His
agricultural knowledge was found worth mentioning in the Vita, the life of Benno,
written by the abbot Norbert of Iburg shortly after Bennos death.
There is no way of knowing what Bennos sources were, whether he was a
clever and practical person who outdid his contemporaries by means of his own
thinking, or whether he had read any of the agricultural manuals of Antiquity,
which were available, if scarce, at his time. Actually, one could also understand the
juxtaposition in the text non usu sed arte in this way: that he was above the
usual in his doing.
Bennos life is part of the unwritten history of agricultural knowledge
which begins in Europe with one of the earliest preserved works of Greek poetry,
Hesiods Works and Days, and continues to the present day. The medieval part of
this history is the least well known, but no detailed knowledge exists for any time
and any part of Europe.
Knowledge about agriculture, of which soil knowledge is part, can be found
from Ancient Greek times onwards in agricultural textbooks, in manuals and agri-
cultural calendars, in texts by natural philosophers and theologians and, later on,
by biologists and natural scientists. Some bits and pieces can also be found in other
texts such as charters, legal texts and administrative (e.g. land taxation) records and
in topographical descriptions which contain information about soils. The Vita of
Bishop Benno is a case in point illustrating the wealth of sources but it also serves
as an example of the difficulties one can expect to encounter in locating them.
Humans living in agricultural societies possess knowledge about how to
produce harvests. They have competence in dealing with domesticated animals
and plants and their competitors and parasites, they are acquainted with various
aspects of their doing, and through trial and error they possess means to recognise
information as being correct; therefore one can summarise that they are in posses-
sion of the three basic constituents of knowledge, competence, acquaintance and
recognition of correctness (Mattey, 2002).
While I shall deal with the concept of knowledge in the next section, and with
the agricultural textbooks in the main part of this chapter, I shall first demonstrate
the context of the more specialised soil knowledge by alluding to the mentality of
agricultural societies. It is commonplace that Greek and Roman mythology show
peoples concern for agriculture, with Demeter (Ceres) as the goddess of agriculture
180
Verena Winiwarter
ever committed any evil with this land. It is significant that this is the end of his
lament. A peasant can go no further than putting his land at stake.
si adversum me terra mea clamat et cum ipsa sulci eius deflent
si fructus eius comedi absque pecunia et animam agricolarum eius adflixi
pro frumento oriatur mihi tribulus et pro hordeo spina finita sunt verba Iob
[if my land cries out against me and all its furrows are wet with tears,
if I have devoured its yield without payment or broken the spirit of its tenants,
then let briers come up instead of wheat and weeds instead of barley.
The words of Job are ended.] Job 31: 3840.
Turning to the New Testament, Matthew 13, 39, Luke 8, 49 and Mark 4,
39 all tell the same story of the difference between sowing on fertile ground and
among thorns as an allegory of the word of God and its reception by the human
soul. An often cited biblical story is the one in Matthew 13, 2431, where not
so much the soil itself, but the problem of weeds serves as an allegory of evil and
its subsequent fate (the weeds are to be collected and burnt). Weeds are used in
allegory also in Hosea 10, 4: loquimini verba visionis inutilis et ferietis foedus et
germinabit quasi amaritudo iudicium super sulcos agri [They make many promises,
take false oaths and make agreements; therefore lawsuits spring up like poisonous
weeds in a ploughed field].
To end up unburied, and be used as manure seems to be one of the worst
threats one can issue, as can be found, for example, in Jeremiah 16, 4:
mortibus aegrotationum morientur non plangentur et non sepelientur in sterquilin-
ium super faciem terrae erunt et gladio et fame consumentur et erit cadaver eorum
in escam volatilibus caeli et bestiis terrae
[They will die of deadly diseases. They will not be mourned or buried but will be like
refuse lying on the ground. They will perish by sword and famine, and their dead
bodies will become food for the birds of the air and the beasts of the earth]. Note
that what is translated as refuse here is stercor, usually translated as dung.
Latin Christian writers in Late Antiquity and in the Middle Ages were
familiar with the Bible and used it in sermons or wrote about its interpretation.
Interestingly, they used several words to describe features of the soil which are not
found in the Bible itself. Latin was the spoken language in large parts of the Christian
Ecumene until the thirteenth century, and one cannot tell from where the authors
took their vocabulary, but whether it was from their reading of ancient works on
agriculture or from everyday spoken language, either way proves the ubiquitous
nature of agriculture in the spiritual life of these writers. Again, I shall give only
very few, very specific examples.
The church father Hieronymus (Jerome) lived c. 350420. He was born
in Dalmatia, had travelled in the Middle East, and then lived in Trier and Rome.
182
Verena Winiwarter
He spent his last years in a monastery he had founded. The most famous thing
about him is his translation of the Bible into Latin, for which he also undertook
philological studies. In addition, he wrote several commentaries on books of the
Bible. In the Commentary on Isaiah, he uses vocabulary that cannot be found in
the Bible, talking about infertile swamps, which are muddy and silty, in contrast
to fertile irrigated fields. The commentary relates to Isaiah 14, 2223, the biblical
text of which is given below.
et consurgam super eos dicit Dominus exercituum et perdam Babylonis nomen et
reliquias et germen et progeniem ait Dominus
et ponam eam in possessionem ericii et in paludes aquarum et scopabo eam in
scopa terens dicit Dominus exercituum (Vulgate)
[I will rise up against them, declares the LORD Almighty. I will cut off from
Babylon her name and survivors, her offspring and descendants, declares the
LORD . I will turn her into a place for owls and into swampland; I will sweep
her with the broom of destruction, declares the LORD Almighty.] (Translation
of the New International Version)
ubi non est ager irriguus, qui afferat fructus diuersorum seminum, sed paludes
infertiles, et limosae ac lutosae, in quibus caeno gaudentia reptant animalia.
[where there is not an irrigated field, which produces fruit from diverse seeds, but
infertile swamps, muddy and silty, where animals are creeping rejoicing in mud.]
(Hieronymus, Commentarii in Isaiam, 6, 14, 23, trans. VW).
None of the terms limosus, lutosus or irriguus are from the biblical text, as one can
see (nor are they found anywhere else in the Vulgate). Hieronymus tries to explain
what one is to understand by the biblical swampland and its owls using terminol-
ogy he must have known from elsewhere.
But Hieronymus was not the only one to use such metaphor. His contem-
porary, Rufinus of Aquileia, likens the care of the soul in all aspects to agriculture,
mentioning ploughing, thorough watering, breaking of the soil (scindere) and fur-
rowing, so that the seed of charity, hope and justice can grow:
Agricola terrae suae putandus est ille, qui campos animae suae cordis que sui noualia
indesinenter uerbi dei aratro et scripturarum uomere scindit et sulcat, qui plantaria
fidei et caritatis ac spei et iustitiae de israhel fontibus rigat et omnem agriculturae
disciplinam in animae suae rure deprendit
[Take him as a farmer of his fields who ceaselessly tears and carves his unbroken land
with the plough of the Lords word and of the Scripture, who irrigates the plantations
of Faith, Love and Hope and Justice from the springs of Israel, and who gathers
all knowledge of agriculture in the lands of his soul.] (Rufinus, De benedictionibus
patriarcharum, II, 14, trans. VW).
183
Prolegomena to a History of Soil Knowledge in Europe
The oikos, the household, as the core unit of society dominated Western
thought for centuries. One should also bear in mind that the only economic oc-
cupation Roman Patricians were allowed to pursue was land ownership. Making
money from money was considered improper or even forbidden in many agricul-
tural societies. Even if such restrictions were often more theoretical than real, the
education of heirs to an estate needed to comprise agricultural matters to ensure
their legal income. A copy of John Loudons famous Encyclopaedia of Agriculture of
1883 (Loudon 1883[1839]) has a dedication that allows us to glimpse the thinking
about proper education of landowners which still existed in the late nineteenth
century:
To Evan Gordon Macpherson [...] on the occasion of his attaining his maturity, With
every good wish from A Macpherson, Kingussie, 15th June, 1889 (the signatory is
probably Alexander MacPherson, who was then provost of Kingussie).
While I have only briefly sketched the importance of agriculture in the
world-view of agricultural societies, what I hope has become clear now is that
with Antiquity, the Middle Ages and Early Modern Times in Europe alike, we are
dealing with societies whose culture is inextricably tied to agriculture. This should
be borne in mind when judging the reception of specialised agricultural texts by a
readership whose literacy is usually underestimated, especially for the Early Mid-
dle Ages (McKitterick 1988), and whose ways of knowing what they were doing
are quite unknown.
ture at a given time and place, but have not studied the evolution of knowledge
as such. Famous historians of agriculture are divided over the question of whether
the agricultural manuals influenced the practice of agriculture. Duby was rather
positive about an influence, whereas Adriaan Verhulst denies it completely (Verhulst
1989, p.1). Even Verhulst admitted that a thorough study of the works would
be necessary to decide about their practical relevance, and in the years since the
publication of his short piece, no such work has appeared. John Ziman has pointed
out that the interrelation of theory and practice cannot be caught by a simple
theoretical model (Ziman 1976, p.35), and the present study, which is meant as
study of ideas about soil, starts by analysing texts rather than aiming directly at
the question of practices.
Within environmental history, the history of ideas has a long tradition.
Clarence Glackens seminal volume Traces on the Rhodian Shore: Nature and Culture
in Western thought from Ancient Times to the End of the Eighteenth Century is a clas-
sic, frequently reprinted and widely used (Glacken 1967). A distinctive feature of
environmental history is its connection to environmental sciences or, more globally,
its interest in being relevant to relations between society and the environment.
Concepts from biology or from environmental sciences have played an important
part in the development of the field. Equilibrium, resilience and adaptation are
prominent examples of biological notions used for environmental history narra-
tives. The notion of sustainability or sustainable development is the narrative
basis for many such studies, which ask if a particular society in a given place was
sustainable or not, and if so, why. The critique of sustainability has also been a
topic for environmental historians (Worster 1988; Winiwarter 2001). This chapter
is an attempt to address an ongoing debate in the sustainability context. In prepara-
tion for the Johannesburg World Summit, The International Council of Scientific
Unions (ICSU) issued a report on Science, Traditional Knowledge and Sustainable
Development (ICSU 2002). Traditional knowledge therein is distinguished from
pseudo-science, and a partnership between science and traditional knowledge is
envisaged to foster sustainable development. The role of written knowledge from
Europe before the so-called scientific revolution, or works that appeared after it
but had no connection to it, is quite unclear in this paper, which does not mention
historical knowledge at all.
In the ICSU paper the boundaries between science, pseudo-science and
traditional knowledge are drawn in a quest to determine the authority of knowledge
(what statements are taken to be true and relevant at a given point). The ICSU posi-
tion is strictu sensu ahistorical, and has to be differentiated in historical perspective.
As the paper shows, questions of epistemology and questions of authority arise in
conjunction with soil knowledge.
This elaboration on soil knowledge history is therefore written as a Foucault-
ian archaeology rather than as a history of ideas. A Foucaultian archaeological
186
Verena Winiwarter
inquiry, in essence, asks what conditions foster particular statements that come to
be taken as true, alluding to the way memory is conceptualised in cognitive theory.
The appearance of one statement rather than another is, of course, not a simple
question for Foucault. A statement is not an isolated utterance, but always belongs
to a series or a whole. [I]t is always part of a network of statements (Foucault
1972, p.99) which compose a discourse. Reflexively, the discursive formation a
relatively autonomous system of serious speech acts in which [a given statement]
was produced (Dreyfus and Rabinow 1982, p.49) sets the context in which
constitutive statements are held to make serious sense, to be true. And truth,
for Foucault, is no more than an ensemble of rules [and] a system of ordered
procedures for the production, regulation, distribution and operation of statements
(Foucault 1980, pp.1323).
Later on, Foucault has worked with the notion of genealogy, which does
not intend to displace the archaeological dig, but to broaden the scope of inquiry.
It seeks to trace the descent and emergence (Foucault 1984) of new discursive
formations; to trace a discourses lineage across the path of contradictions and logical
discontinuities the accidents, chance, passion, petty malice, surprises and power
(Davidson 1986, p.224) that foster new discursive formations (Rice 1992).
Foucaults emphasis on the non-linear character of knowledge as produced
in discursive formations is important as a concept because it works against the false
impression of a body of accumulating knowledge, its quality improved by succession
towards truth (which would then mean: current scientific concepts). Secondly, he
emphasises that writing history is an act of rewriting it, (re-criture) rather than
an act of reconstruction: the history I offer here is my rewriting, and should be
understood as part of the discourse rather than as outside of it. Especially if one
deals with pre-modern times, the vagaries of the archival situation, the availability,
visibility and invisibility of parts of the body of knowledge for authors who wanted
to contribute to these discourses, should be taken into account when judging the
relationship of a text to the existing contemporary literature. If, for example,
Petrus de Crescentiis, whom I shall present later, did not cite Columella directly,
but only Palladius, this could mean that he did so by choice, or it could mean
that only Palladius was available to him. Many such questions will unfortunately
remain without answer.
Franois Jacob has written a fascinating account of the development of
biology, making excellent use of Foucaults concept of an archaeology of knowledge
(Jacob 1972). According to Jacques LeGoff, Jacobs book can be considered the
only successful archaeology of knowledge he had ever encountered (LeGoff et al.
1990, p.42). How the aims of such a project differ from a type of historiography
where the history of knowledge is taken as factual, has been explained by Athar
Hussain: Archaeology does not seek to trace the intentions, motives, etc., of the
speaking subject. It is not the discovery of the primordial germ of a discourse. It
187
Prolegomena to a History of Soil Knowledge in Europe
problem of defining what data are. The other dictionary definition, information
being an accretion to knowledge is circular. Information will therefore be defined
as data (observation-based impressions on the human senses) which under given
circumstances are considered as different from noise (i.e. all the impressions on the
senses which are not used in interpretation of the world from which they emanate).
After this excursion into information theory, it should have become clear that
knowledge rests on a set of decisions made by humans on how to proceed with
sensory perceptions. It is such decision-making processes that interest me as histo-
rian of knowledge, though without any hope of being able to study them directly.
A second order type of knowledge-generating process can be studied, however:
the web of communication within which decisions about noise versus data, about
interpretation of data and about strategies of communication itself are made and
can be traced or at least re-written by the historian.
The next question is about the relationship between knowledge and experi-
ence. One should note that experience and experiment both come from the same
Latin word, experior, a verb meaning try or, as it still means in English today,
experience something. Experimentum, also a Latin word, accordingly means a
trial or test.
Experience, according to dictionary wisdom, is an activity that includes
training, observation or practice and personal participation. In the dictionary
the notion bears a strong connection to the life-world, as experiences are lived
through or happen to one (Webster 1983, p.645). If one accepts this definition, the
process(es) by which experience is transformed into knowledge remain(s) unclear.
The transformation of experience into language, be it oral or written, is critical and
needs attention. In such a context, the notion of tacit knowledge, which is used
to describe the unspoken or implicit in an action which the actors are unable to
formulate in language, might be helpful. Another distinction along these lines it
the one between embodied knowledge and theoretical knowledge. Knowing-how
or embodied knowledge is characteristic of the expert, who acts, makes judgments
and so forth without explicitly reflecting on the principles or rules involved. The
expert works without having a theory of his or her work; he or she just performs
skilfully without deliberation or focused attention. Knowing-that, by contrast,
involves consciously accessible knowledge that can be articulated and is charac-
teristic of a person learning a skill through explicit instruction, recitation of rules,
attention to his or her movements, etc. While such declarative knowledge may be
needed for the acquisition of skills, the argument goes, it is no longer necessary for
the practice of those skills once the novice becomes an expert in exercising them,
and indeed it does seem to be the case that when we acquire a skill we acquire a
corresponding understanding that defies articulation. That an exhaustive equation
of tacit knowledge with pre-theoretical, skilled expertise cannot be maintained
189
Prolegomena to a History of Soil Knowledge in Europe
becomes particularly clear when we consider that one widely accepted paradigm
of tacit knowledge is to be found in language competence. In contrast to other
varieties of tacit knowledge, knowledge of language is not understood to constitute
a skill, and thus to consist in a capacity to do something but rather is a properly
cognitive capacity, and therefore defined in terms of mental states and structures
that are not always or reliably manifested in behaviours or performances. There
is thus reason to suppose that at least some though by no means all forms of
tacit knowledge can behave like ordinary dispositions to believe, and accordingly
can be brought to awareness given the proper circumstances. We might say then
that these kinds of tacit knowledge are tacit to the extent that they are initially
inaccessible to the person to whom they are attributed, but that given the proper
conditions, this inaccessibility can be converted to the kind of accessibility enjoyed
by our ordinary knowledge (Barbiero n.d.).
In the transformation of (tacit) knowledge from practical experience and
from the ubiquitous metaphorical context found in agricultural societies into a
specialised agricultural text (whether a manual or a poem, if its theme is agricul-
ture it will be considered an agricultural text), it is changed and reformed in the
conversion, and becomes part of a body of texts dealing with agriculture, texts with
diverse aims and audiences, written from different experiences and by authors with
different skills in the transformation process described above.
The theme of their works, agriculture, nevertheless poses similar writing
problems for all of the authors. These are reflected in similarities in their texts. But
the notion of similarity is not a precision tool. To formulate the dissection plan
in Foucaults terminology, all those texts should be considered as belonging to
one discursive field. Within it, one needs to differentiate between several forms of
difference/similarity: linguistic analogy (which means the ability to be translated),
logical identity (equivalence) and homogeneity of the statement. Changes in each
of these three dimensions can be indicative of the formation of a new discursive
practice, and it is such discursive practices that I will try to distinguish from each
other. But synchronic and diachronic comparisons will be needed for that, as
several discursive practices can co-exist at the same time. I shall not analyse the
entirety of the works I refer to, but confine myself to those parts that are closely
linked to soils.
While I consider it important to distinguish between knowledge, practice,
experience and the like, in analysing agricultural treatises one has to bear in mind
that the differentiation we make today between practice and theory and between
experience and experiment are relatively new (post-Baconian or even post-eight-
eenth century) developments, at least in the English language. Therefore, one must
be careful not to misinterpret these two synonyms as being different from each
other in any particular way when one finds them in the sources.
190
Verena Winiwarter
Table 7.1. A selection of extant agricultural treatises of classical antiquity and beyond
All extant works mentioned in this text can be found in Table 7.1, which is ordered
according to the age of the texts. Many more must have existed, if we trust the
authors who mention many names of experts of whom this reference is the sole trace
we have. One of the most mysterious works is a treatise written by a Phoenician
named Mago. As he is sometimes called the father of agricultural textbooks I find
it important to show that the reference to his work is reference to an unknown.
We have only very blurred traces of it but to follow them is extremely interesting.
Magos work is said to have comprised 28 volumes. It was rescued during the fall
of Carthage (146bc) so it must be older than that, and the Roman Senate even
commissioned its translation, as Pliny the Elder reports in the book on agriculture
in his encyclopaedia, where Mago is mentioned alongside with other non-Latin
writers:
Igitur de cultura agri praecipere principale fuit etiam apud exteros, siquidem et
reges fecere, Hiero, Philometor, Attalus, Archelaus, et duces, Xenophon et Poenus
etiam Mago, cui quidem tantum honorem senatus noster habuit Carthagine capta,
ut, cum regulis Africae bibliothecas donaret, unius eius duodetriginta volumina
censeret in Latinam linguam transferenda, cum iam M. Cato praecepta condidis-
set, peritisque Punicae dandum negotium, in quo praecessit omnes vir clarissimae
familiae D. Silanus.
192
Verena Winiwarter
[To give instruction on agriculture was a matter of importance also for foreigners,
because even kings dealt with it, such as Hieron,2 Philometer Attalos,3 Archelaos4,
and military leaders like Xenophon, and even Mago of Carthage, whom our senate
paid such respect after the capture of Carthage, that while it distributed the library
among the kings of Africa, it had only the 28 books of this man translated into Latin,
although M. Cato had already written such a work. The senate confided this work
to men experienced in the Punic language; among whom D. Silanus, a man from a
distinguished family, stands out] (Plin. n.h. XVIII, V, 22f., trans. VW).
Two of the prominent Roman writers on agriculture, Varro and Columella,
inform us about the subsequent history of Magos work. Varro gives an account
that shows great interest in Magos work and hints at the fact that the translator
was not in fact a mere translator, but added Greek knowledge and left out part of
the original work.
All these are surpassed in reputation by Mago of Carthage, who gathered into 28
books, written in the Punic tongue, the subjects they had dealt with separately. These
Cassius Dionysius of Utica translated into Greek and published in twenty books,
dedicated to the Praetor Sexti(li)us. In these volumes he added not a little from Greek
writers whom I have named, taking from Magos writing an amount equivalent to
eight books. Diophanes, in Bithynia, further abridged these in convenient form into
six books, dedicated to king Deiotarus. (Varro, rust. 1,1,10, transl. Loeb Classical
Library, pp.165, 167)
Varro compresses a history of about 100 years into these few sentences. Carthage
was destroyed in 146bc, and the translation was made soon thereafter, around
140bc. The abridged six-volume edition can be dated to around 60bc because
one can date King Deiotarus to this time. The Greek translation was made in
between, and it has been suggested that it was commissioned in connection with
the colonisation projects of C. Gracchus, because in several of them Greek would
have been the only common language of the people assembled from different parts
of the Empire (Mahaffy 1889/90, pp.33f ).
Columella, who refers to the same incident, is far less precise: As a matter
of fact, Diophanes of Bithynia epitomised in six abridged volumes the treatise of
the entire work of Dionysius of Utica, who translated in many prolix volumes the
treatise of the Carthaginian Mago (Col. r.r. 1, V, 10, transl: Loeb Classical Library,
Col. r.r. Book 1, p 33).
Neither the translation Columella refers to nor its abbreviated version have
been passed on to us. So, unfortunately, we have no knowledge of the scope of
this work, but Magos book was obviously important to the authors we do know
2
Tyrant of Syracuse
3
King of Pergamon
4
King of Macedonia
193
Prolegomena to a History of Soil Knowledge in Europe
of. Mago is of particular interest because through the references to his work the
geographical extension of the Mediterranean world in which agricultural knowl-
edge was exchanged can be anticipated. The active interest by the Senate is a very
important proof of the high value agriculture had, and the complicated history of
translation, expansion and compression makes visible the discursive formation to
which these authors belonged.
The oldest works (also older than Magos treatise) mentioned in Table 7.1
are by Greek authors. Hesiod predates all others by centuries. In his poem (there
was no prose writing in the sixth centurybc) he placed agriculture in the context
of myth. He recounts the generations of beings created by the gods, humans being
the fifth generation. He then goes on to describe the works that have to be done
in the yearly cycle, for which the stars serve as guides. The rise and setting of the
Pleiades and of Arcturus are used as indicators of time for sowing, reaping or other
work, the themes being organised according to an astronomical calendar. His list
of things to do does not include any reference to specific types of soil. But Hesiod
is very specific about the type of wood best suited for constructing a plough; he
recommends having a spare plough in case one breaks; he gives details about the
age of oxen best suited to do the work and how to go about it; and he has a lot to
say about the right timing of ploughing, planting vines, cutting wood and other
agricultural tasks. He generally recommends getting new land under the plough
enlarging the agricultural area was a good thing to him.
The other important Ancient Greek treatise, Xenophons, is the first extant
example of another narrative structure, the form of a dialogue between Socrates,
Kritobulos and others, among them the landowner Isomachos as an expert. The
Oikonomikos, as this book has been called, has some information on soil qualities,
albeit not very differentiated. The basic idea is that soil and plant are an interac-
tive system: For they tell us that to be a successful farmer one must first know the
nature of the soil. Yes, and they are right, I remarked; for if you dont know what
the soil is capable of growing, you cant know, I suppose, what to plant or what
to sow (Xen., Oikon., 16, 2). Soil can be fat or lean (17, 8), dry or moist (e.g.
16, 11), and by means of the plants that grow on a piece of ground one can judge
the quality of this ground (16, 3) (such plants are called indicator plants today).
Xenophon, like Hesiod, is concerned about the right timing for ploughing and
sowing (17, 5). Manuring is known as a means of improving the soil. Some editors
hold that particular reference is made to green manure (Oikon., 16, 12), as is the
application of salt-free, dry or wet substances to clean soil from salt (20, 12), a form
of land improvement. The multifaceted knowledge of agriculture does not easily
yield to narration, and Xenophon uses the form of the dialogue in a very elegant
and convincing way to convey his information. However, Xenophon is mainly
concerned with the economic output of the farm and with its management, and
194
Verena Winiwarter
less with details of agricultural operations. We might consider his rendering of soil
quality as the most basic knowledge one can expect in agriculture.
Hesiods and Xenophons are the two most widely known Greek agricultural
works; the third one that is widely known is a Byzantine collection of older works
(which we only know via this collection) called Geoponika. Much discussion has
been going on about this collection, which was apparently compiled in the tenth
century ad, but comes from works dating back at least to the sixth century. Most
probably it was compiled by an unknown writer for the Emperor Constantine VII
Porphyrogenitus, to whom the work was formerly ascribed. It is based mainly on
a collection made in the sixth or seventh century by Cassianus Bassus, who used
two earlier writers, Anatolius Vindanius (or Vindonius) of Berytus, and Didymus
of Alexandria (both of the fourth or fifth century). The most recent and excellent
summary and discussion can be found in a study on Byzantine gardening (Rodg-
ers 2002). In his study Rodgers also shows the close connection between Latin and
Greek and Arabic texts (see Figure 7.1), again pointing to the cross-fertilisation of
several traditions, such as we have already seen when discussing Magos work.
Due to their expansion from the seventh century onwards, Arabs became
a major political and cultural factor in the Mediterranean world. Their flourishing
realms on the Iberian Peninsula became cultural contact zones of great importance.
There is at least one famous Arab treatise on Agriculture (Ibn Al Awwms Kitab
al-filaha). We know next to nothing about the author, who was an Arab from Al
Andalus and wrote his treatise in the middle of the twelfth century. He compiled his
work using agricultural and veterinary medicine works of Greek and Latin origin,
which he presumably had available in Arab translation, putting them together without
much comment. But selection is also a creative process of knowledge generation,
and as Karl Butzer (1994 and 1993) has shown, he is part of the multi-lingual
network of Mediterranean agricultural authors who all knew many others works
and used them according to their own selection criteria and preferences with their
own experience as a guideline for selection. Magos work was translated on behalf
and for the Roman Senate. The Geoponika was another work of public interest, as
far as the interest of the aristocracy of mid-tenth century Byzantium can be called
public. The overall value placed on agriculture can be seen in the preface to one of
the manuscripts of the Geoponika:
Knowing that the state consists of three elements the army, the clergy, and agri-
culture you have devoted no less care to the latter, which is best able to preserve
human life. That which several of the ancients discovered through their study and
experience of cultivating the land and their care of plants, the seasons, the meth-
ods and terrain suitable to each, and furthermore the discovery of water and the
construction of buildings, their implantation and orientation, all these and many
other important things, the greatness of your genius and the depth of your spirit
have gathered together, and you have offered to all a work that is generally useful.
Khpopoii?a: Garden Making and Garden Culture in the Geoponika 163
195
Prolegomena to a History Table
of Soil
1
Knowledge in Europe
Literary Traditions of Agricultural Writers
(/ ..)
e.g.: Mago the Carthaginian, Pseudo-Democritus
e.g.: ar-Razi (IX), Ibn Haggag (XI), Ibn al-Awwam (XIII)
Figurewhich
7.1. The connections
this manuscript doesbetween
preserve. texts
Aside on
fromagriculture
the absence in Arab,
of the Latintheand
prologue, Greek,
overall
text of the Marcianus,according
give or taketotries here (2002,
Rodgers and there, is the same as other Geoponika
p.163)
manuscripts: signicantly it includes what are apparently Constantinian features, such as
the chapter on the growing season for vegetables in the area of Constantinople (Geopon.,
12.1), and mythologies associated with certain plants in chapters of book 11. Second, the
Arabic versions
Immediately, he derived from Cassianus
who applies (both
himself to theKassianos andlabours
fruit of your Qust.usisinable
Table
to1)recognise
reveal an
exactly what his existence consists of, and he may observe in perfect order that which
is both useful and necessary, what the basis of human life is and that on which he
lavishes all his care. He can see not only what is necessary, but also superfluous things,
conducive solely to the enjoyment of his eyes and his sense of smell. (Prooimion
69, trans. Lefort 2002).
Like the above-mentioned works, the Geoponika also contains information
on soils, e.g. in Book 2, Chapters 911 and in Book 12, Chapter 3, which talks
about soils for vegetables. In Chapter 12, 2 we encounter soils in the garden:
196
Verena Winiwarter
(12, 2: 3) The person who wants to excel in growing garden plants must take fore-
thought for good seeds, suitable soil, water, and manure. (4) Good seeds will produce
offspring like themselves. Suitable and fertile soil will guard what is entrusted to
it. Water will make the vegetables grow larger through nurture. Manure makes the
soil more friable, so that it receives water more readily, to make space for the roots
and to allow the foliage to sprout (trans. Rodgers 2002, p.173).
While today we have another theory as to why manure works, the passage contains
some interesting conceptual remarks on soils. That there is fertile and infertile soil
was probably the most common knowledge about agriculture, and that soil and
plant have to match to yield good results was also standard wisdom. Friability as
a distinct soil quality is interesting in terms of soil knowledge.
The Geoponika has been widely translated into, among other languages,
Arabic, Latin, German and French (in 1545), but no modern translation of the
whole work is available. Rodgers at least has a table of contents, which is helpful
(Rodgers 2002, p.166).
In Figure 7.2, I have tried to convey that when dealing with agricultural
writers we are dealing with a synthetic category which encompasses works whose
theme might be (in part or in general) agriculture, but which are part of several
different genres. I have to emphasise that my categories are haphazard (as all such
schemes are). Connections exist from within the agricultural literature to other
realms of knowledge. Whereas I have treated works on (home) economics such as
Xenophons as part of the agricultural realm, I have excluded important sources for
the agricultural tradition, especially for the learned treatises, viz. botanical books.
The most important and influential botanist of antiquity was Theophrastus (371
286bc), a pupil of Aristotle. He was also in charge of the first existing botanical
garden. Two of his works still exist: Historia plantarum (A History of Plants) and
De causis plantarum (On the Reasons of Vegetable Growth). Quite often, the way
an ancient text is classified and studied is determined more by modern scholarship
than by the distinctions drawn by the texts authors themselves. A few examples
of the agricultural tradition will elucidate this point. Albertus Magnus, although
he deals with agriculture as much as other compilers of encyclopaedias did, has
mainly been encompassed into the history of science, and not into the history of
agriculture. The famous quotation which makes him the predecessor of scientific
rigour, Experimentum solum certificat in talibus (Experiment is the only safe
guide in such investigations, De vegetabilibus, VI, tr. ii, i), actually comes from
his botanical treatise, which might as well be considered part of his agricultural
studies. As mentioned above, at the time he was writing, no sharp distinction
existed between experiment and experience, so the historians of science might be
overstating their case. In contrast to him, Petrus de Crescentiis, who draws heavily
on the work of Albertus, is counted among the agricultural texts. Historiography
has its mantraps.
197
Prolegomena to a History of Soil Knowledge in Europe
AGRICULTURAL
T EXTBOOKS
a) Notebooks
POEMS LEARNED TREATISES Cato, De agri cultura
Hesiod, Works and Days dealing with Grnthaler, Haushaltungsbchel
Virgil, Georgica AGRICULTURE b) Systematic Treatises
Columella, De re rustica, Varro, Res Rusticae Columella, De re rustica
Book 10, On Gardening Petrus de Crescentiis, 1, 9, 11, 12
Palladius, Carmen de Insitione Ruralia Commoda Konrad von Heresbach,
Wahlafrid Strabo, Albertus Magnus, Rei rusticae libri Quatuor
Liber de cultura hortorum De vegetabilibus et plantis
HAUSVTERLITERATUR
CALENDARS ENCYCLOPAEDIAS Conrad von Heresbach,
Menologiae Rusticae Pliny, Naturalis Historia, Rei Rusticae Libri Quattuor
Palladius, Books 3-13 esp. Book XVIII Coler, Oeconomia
Wandalbert of Prm Isidor of Seville, Etymologiae, Martin Grosser, Anleitung
Carmina Chapter XVII, zu der Landwirtschaft
Coler, Calendarium Hrabanus Maurus, De rerum Abraham v. Tumbshirn,
oeconomicum naturis, Chapter 19 Oeconomia
& perpetuum Konrad von Megenberg, etc.
Ykonomica
Vincent of Beauvais, ECONOMIC TREATISES
SPECIALISED BOOKS
Speculum maius, liber sextus Xenophon, Oikonomikos
Gargilius Martialis,
John Loudon, Walter of Henley,
De arboribus pomiferis
Encyclopaedia of Agriculture The Treatise on Husbandry
Palladius, De arboribus
Further Books on COLLECTIONS
Horticulture, Arboriculture Geoponika
Along the same lines, the theological interests of Hrabanus Maurus are
certainly the most important aspect of his work, but it would certainly be wrong
not to treat him also as someone who used agricultural knowledge. So Figure 7.2
is also an attempt at bringing together works that have been studied under quite
different presuppositions, treating them as the precursors to economy (Xenophon),
natural science (Albertus Magnus), bucolic poetry (Virgil), etc., and not trying to
compose a larger picture of the many ways in which agricultural knowledge was
present in these works and thus in learned circles.
These historiographic considerations have brought us to the beginning of the
fourteenth century ad. Now we must go back 1600 years, to the second centurybc.
The first prose work ever written in Latin was an agricultural notebook. This fact is
another sign of the importance of agricultural knowledge in solar-based societies.
198
Verena Winiwarter
Cato the Censor, a distinguished Roman patrician, left a work about agriculture
which has more the character of a journal than of a systematic treatment of the
subject. But the author is aware of the importance of appropriate timing, saying,
Opera omnia mature conficias face. Nam res rustica sic est, si unam rem sero feceris,
omnia opera sero facies [See that you carry out all farm operations betimes, for this
is the way with farming: if you are late in doing one thing you will be late in doing
everything.] (Cato, agr. V, 7). Cato, however, does not give any kind of calendar of
tasks. A century after Cato another distinguished Roman patrician, Varro, wrote
the treatise Res Rusticae (On Agriculture) in the style and form in which a learned
man of that time would write, as a dispute among three estate owners discussing a
variety of subjects. He was very concerned with etymological aspects of agricultural
vocabulary, a part of his interest we have come to belittle. The book was written
late in his life, and was meant as a manual for his young wife, Fundania, so that she
would be able to oversee the estates after his death. This might be purely narrative
construction yet it is interesting, because Varro understood his book as practical
advice. The work consists of three books: one on animal husbandry, one on ag-
riculture proper, and one on all types of horticultural activities and the breeding
of fowl. In the first book he talks about the two measures of time one being the
annual solar cycle, the other being the monthly lunar cycle and he elaborates at
some length the various periods into which the year can be divided. (Varro, rust. I,
XXVIII f.). First, Varro divides the year into eight periods and explains what should
be done in each, then he talks about the lunar cycle. His general remark that there
is a planting season appropriate for each kind of seed shows that he, too, is aware
of the importance of timing (Varro, rust. I, XXXIX). Varro makes an interesting
remark about the grafting of plants, saying that some that were formerly grafted
in spring are now grafted in summer. Obviously there was some development of
agricultural techniques of which Varro takes account (Varro, rust. I, XLI).
Agriculture played a very important role in ancient Roman life; high-rank-
ing individuals were concerned with and about it. So, it is no wonder that even a
work of verse was written by a very prominent Roman poet, Virgil. His Georgics
are again divided into books dealing with the various subjects of agriculture, as
is Varros work. In the first book, Virgil explains the main tasks of fieldwork. The
second part of the first book (204310) is an agricultural calendar, and the third
part is a meteorological treatise. His second book is about trees, the different
kinds of soils and climates, and the work to be done in an orchard. The last part
of this book is in praise of rural life. The third book is about both large and small
animals, and the fourth is about bee-keeping, beginning with the legendary crea-
tion of the bees and continuing with a dramatic mythological story, in which no
practical agricultural knowledge is presented. Virgil is unsurpassed in the amount
of manuscripts that have survived. He was certainly very widely known in later
times. He also must be credited with being the founder of the agricultural poem,
199
Prolegomena to a History of Soil Knowledge in Europe
in which he had two important followers. The books by Columella and Palladius
each contain a part in verse. Columellas tenth book on gardening and Palladiuss
Carmen de insitione (On Grafting) were written (as the authors let us know) in
homage to the Georgics. The Hortulus of Walafrid Strabo, a poem on the plants
of the garden written by the learned Abbot of the Reichenau monastery (formerly
one of the teachers of Charlemagne), is an early medieval example of this literary
tradition, which has had importance in maintaining the status of agriculture as
part of culture ever since.
By modern standards, the most scientific ancient work ever written on
agriculture in pre-industrial times is surely Columellas treatise on agriculture (De
re rustica). We do not know much about him. He lived in the first century ad, was
born in Spain and passed through a typical Roman career. He had several estates, and
as we can infer from his invention of a new implement for grafting of trees, he must
have taken part in the practical running of them. His agricultural work was written
late in his life. It consists of twelve books, amongst which the one on horticulture is
written in verse, as mentioned above. The other books are organised by topics; one
entire book (Book 2) is on soils, but the work also contains an agricultural calendar
using astronomic markers (e.g. the rise of the Pleiades) to give dates.
Within the socio-political entity of the Imperium Romanum, several more
specialised treatments of agricultural matters were undertaken, and agriculture
also became part of the encyclopaedic attempts. We know of specialised works
on viticulture only by way of their mention by Columella, but several treatises on
veterinary medicine and at least fragments of a work on arboriculture written in
the third century ad survive. While veterinary medicine will not be dealt with here,
Gargilius Martialis work on tree culture, which contains detailed information on
soils, is an important part of the discursive formation. Unfortunately, we know
nothing about the author of this book.
Pliny the Elder, who was not primarily an agricultural writer, did write the
very famous encyclopaedia Naturalis Historia, summing up the knowledge of his age,
including in Books 17 and 18 a great deal of agricultural knowledge and a calendar
of agricultural tasks. Pliny, his Latin full name Gaius Plinius Secundus, was born
in ad23 in Novum Comum, Transpadane Gaul, and died on 24 August ad79 in
Stabiae, near Mount Vesuvius, while studying the eruption of the volcano. Pliny
came from a prosperous family, and could pursue studies in Rome. He embarked
on a military career, serving in Germany, and on his return to Rome he possibly
studied law. He later became procurator in Spain. His devotion to his studies and
his research technique were described by his nephew, Pliny the Younger. Upon
the accession of Vespasian in ad69, Pliny returned to Rome and assumed various
official positions.
We know of another encyclopaedia, by Aulus Cornelius Celsus, of which
only the medical part survived. His work also contained chapters on agriculture.
200
Verena Winiwarter
work on which to base their own thinking. One often forgets that the conscious
use of expert knowledge is a creative act, and in this case in particular, one can see
that the choice was made according to a quite modern principle, taking the latest
and most advanced work.
Konrad of Megenberg, the last person to introduce, is again a cleric. He, like
so many others I have mentioned, travelled through Europe and was actively engaged
in the politics of his day. His work encompassed the Yconomia, an encyclopaedic
treatment of all possible branches of home and public economics. It includes a
lengthy passage on soil treatment and gives enormous detail on soil qualities, and
it is also mainly based on Albertus Magnus. Although Vincent of Beauvais made
use of the work of Albertus, his treatment of agriculture is mainly based on Pal-
ladius, abridged and blended into the larger concept of his Speculum. Palladiuss
second book, which is entirely on soils, is cut to almost half the original length.
This must have been a conscious decision, and is worth studying in detail. Some
of the authors mentioned have recently been treated thoroughly in the context of
literary studies in a dissertation on Latin agricultural writers (Rex 2001). None
of the medieval works exists in a modern translation, although all are edited and
thus available for study.
Of all the detail on soils these works contain, I can only mention very few
examples here. Albertus Magnus explains why one should plough thoroughly, be-
cause the nutrient content of the field increases in the lower layers. This is one of
the first systematic mentions of the multi-layered nature of the soil (but see Col.
r.r. II, 2,21, who mentions the layer below the humic upper soil as important),
and also of the understanding that nutrients are distributed unevenly. For Kon-
rad of Megenberg, the care of the fertility of fields is like the care of the medical
doctor for a patient. He uses this metaphor when he states in the chapter on land
improvement that some kinds of soil cannot be improved: Sed terra sicca salsa
et amara numquam accipit medicinam [But a dry, salty and bitter earth does not
ingest medicine] (trans. VW).
Table 7.2c Latin soil terminology compiled from agricultural works in antiquity
Nouns of general content
ager, agri, m.....................................................field, plot
arvum, i, n....................................................arable land
humus, i, f.................................................. soil (humus)
gla(e)ba, ae, f...........................................................clod
solum, i, n................................................................. soil
terra, ae, f...............................................................earth
tractus, us, m. .........................................................land
Nouns to specify types of soil (or related matter, such as fertiliser)
argilla, ae, f.............................................................. clay
carbunculus, i, m.............................................carbuncle
cinis, eris, m.............................................................. ash
creta, ae, f.............................................................. chalk
limus, i, m.............................................................. mud
limus levis....................................................... cow dung
marra, marga, ae, f.................................................. marl
sabulum, i, n................................................ sand, gravel
stercus, oris, n......................................................manure
uligo, inis, f..........................................................swamp
White has sorted the vocabulary into the categories mineral content, mois-
ture or dryness, structure and texture, warm and cold, richness or leanness and
heavy or light. My suggestion in Table 7.2a is to sort them in a slightly different
way. I have tried to build categories compatible with qualities of soil which modern
soil science discerns. Therefore, I have compiled all words that are about grain
size, density and structure, humidity and colour, which correspond to modern
categories, and fertility, taste, temperature and special properties, which do not,
or not directly translate into modern categories
Columella set up a systematic soil classification system, which rests on
dichotomies: soil can be dry or wet (siccus vel umidus), dense or loose (spissus vel
solutus), and fat or lean (pinguis or macer). His interest is the overall soil quality for
agriculture, and for such an assessment these three qualities are surely encompass-
ing and accurate.
He did not use the systematic treatment of different minerals, as can be
found in Varro. This is remarkable, and clearly shows that decisions about classifi-
cation systems are fundamental and should not be taken for granted. Varro uses a
theory about soils which holds that the different types are generated by mixing of
eleven kinds of (mineral) substances. The passage in Varro (rust., I, 9. 23) reads:
In illa enim cum sint dissimili vi ac potestate partes permultae, in quis lapis, mar-
mor, rudus, arena, sabulo, argilla, rubrica, pulvis, creta, glarea, carbunculus, id est
205
Prolegomena to a History of Soil Knowledge in Europe
quae sole perferve ita fit, ut radices satorum comburat, ab iis quae proprio nomine
dicitur terra, cum est admixta ex iis generibus aliqua re, cum dicitur aut cretosa sic
ab illis generum discriminibus mixta.
[For there are many substances in the soil, varying in consistency and strength, such
as rock, marble, rubble, sand, loam, clay, red ochre, dust, chalk, ash, carbuncle (that
is, when the ground becomes so hot from the sun that it chars the roots of plants);
and soil if it is mixed with any part of the said substances, is e.g. called chalky, as
well as according to other differences as mixed] (translation compiled from Loeb
Classical Library and the German translation by Flach).
Note that sabulum is translated as loam in Loeb, whereas Flach has potters clay.
Translation is interpretation, as I have said before. Besides, the Roman authors
were not trying to be consistent with all their predecessors, and each work has its
distinct use of words. As another example, we cannot judge for sure, if cretosus and
creteus are synonyms, and overall, the difference between clayey and chalky seems
to be blurred.
Temperature is not to be mistaken for degrees Fahrenheit or Centigrade.
A hot soil for Columella should not be given too much manure, as it would be
burned, whereas a cold soil needs a lot of it. The Roman scholars used words derived
from Greek philosophy, where hot and cold were properties not simply denoting
temperature, but more the temperament, as of a human being. A cold soil then
has to be understood more like one of cold temper, one with too much phlegm,
a phlegmatic soil. Of course manuring means adding chemical energy to a soil, so
we can understand Columella today using the energy concept. Whether he himself
had a concept of manure as added energy, one cannot say.
Of the special properties, cariosus will be discussed in more detail. The idea
of the notion is clear for Columella, a practitioner, and he is able to explain it to
us, unlike his predecessors, above all Cato, who only warns you to touch it with
no explanation at all. Terra cariosa is not a soil, but a state of soil which can arise
from inappropriate treatment. Soil that is worked when the upper layer is wet from
rain and the lower layers have not been humidified becomes cariosa, it loses its
fertility for years through ploughing under adverse conditions.
Word and concept come from the rustici the rural people as Columella
informs his readers, and are probably understood by those scholars that have practi-
cal experience. The various authors are more or less successful in explaining it a
typical case of tacit knowledge. It is not understood by Pliny, the encyclopaedist,
as can clearly be seen in his text. He misunderstands terra cariosa as a type of soil,
whereas for Columella it is a result of poor treatment and a (reversible) condition
(Winiwarter 2000).
One must not forget that at least 900 years (from Palladiuss time onwards)
lie between Petrus de Crescentiis and his ancient predecessors. While some of the
words given in Table 7.2b have to be considered as mirroring developments of
206
Verena Winiwarter
the Latin language (such as uliginosus), and some are mere synonyms, some, such
as melencolicus or subcaeneus give proof of new theoretical concepts. It would be
interesting to trace the first appearance of, for example, ager melencolicus, a descrip-
tion of soil quality which can only be understood in the context of humour theory,
meaning a cold and dry complexion. Petrus de Crescentiis wrote for a literate audi-
ence and his book, in my opinion, is an attempt at developing agriculture into a
truly scientific undertaking. He uses the (impractical) alphabetical classification
system, which he takes from the best and most recent scientific work he can get
hold of, and works hard to press the different qualities of soils into the fourfold
scheme of the humour-theory. So wet and dry, and cold and hot have to be seen as
meta-categories, under which the more detailed, practical denominations come to
be subsumed. A complete comparative vocabulary of soil terminology by author as
well as a thorough study of the influence of humour theory on soil concepts could
yield a more detailed picture of the evolution of conceptual knowledge. Such work
is planned but will have to be done in years to come.
Ideoque maximos quaestus ager praebet idem pinguis ac putris, quia cum plurimum
reddat, minimum poscit; et quod postulat, exiguo labore atque impensa conficitur.
Praestantissimum igitur tale solum iure dicatur.
[And on this account, a field which is both rich and mellow (crumbly, VW) yields
the greatest returns, because in producing most it demands least, and what it does
require is supplied with trifling labour and expense. Such a soil may therefore with
justice be called the very best] (trans. Loeb Calssical Library, Vol. I, p.111).
Similar assessments can be found in all Roman authors. In Walafrid Strabos ninth-
century ad work on gardening, general descriptions of good and bad land open
the poem:
Ruris enim quaecumquae datur possessio, seu sit
Putris, harenoso qua torpet glarea tractu,
Seu pingui molita graves uligine foetus
Collibus erectis alte sita, sive jacenti
Planitie facilis clivo, seu vallibus horrens,
Non negat ingenuos holerum progignere fructus,
si modo non tua cura gravi compressa veterno,
[Whatever kind of land you might own, be it
that on crumbly sand infertile gravel it might rest
be it that from fat humidity it might yield heavy fruit
high up on towering hills, on the gently sloping plain or furrowed with valleys
nowhere does it refuse to produce its peculiar plants
if only your care does not exhaust itself in laming laziness.]
(translation with changes after the German adaptation by Stoffler, VW)
While Columella uses putris as a notion for a good piece of land, Strabo has it to
describe the looseness of sand which will fall apart in crumbles and not stick together.
This example shows that a general Latin soil vocabulary is not possible. One will
always have to account for the specific way each author uses the words.
The plant-relative description system will be demonstrated using the Chestnut
tree, for which we also have Gargilius Martialis as a reference.
Ea pullam terram et resolutam desiderat; sabulonem humidum vel refractum tofum
non respuit; opaco et septentrionali clivo laetatur; spissum solum et rubricosum
reformidat.
[It likes a black and loose soil: does not refuse a damp, gravelly soil or crumbling
tufa; delights in a shady slope with a northern exposure; and fears a heavy soil that
is full of red ochre.] (Col. r.r. 4, 33, 2, trans. Loeb Classical Library, Vol. I, p.457).
Note: The Loeb edition translates pullus as black here, I would prefer dark.
quaerit solum facile nec tamen harenosum, maximeque sabulum umidum aut car-
bunculum vel tofi etiam farinam, quamlibet opaco septentrionalique et praefrigido
209
Prolegomena to a History of Soil Knowledge in Europe
situ, vel etiam declivi. recusat eadem glaream, rubricam, cretam omnemque terrae
fecunditatem. (Plin. n.h. XVII, xxxiv, 147)
Solum castanea desiderat molle et subactum; non tamen arenosum patitur et sabu-
lum. Si umor adfuerit, pulla terra vel maxime prodest: sed et carbunculus, itemque
tofus optime servit diligenter infractus. Spissum rubricosum locum non vult: vix
provenit in argilla: in glarea cretaque nec nascitur. (Gargilius Martialis, De Hortis,
IV. 5, 314318)
Sie bekommen gerne auff steinigten festem Lande / und an kalter lufft / doch kann
ihnen warme lufft auch nicht viel schaden. Auf festem und lehmigen Lande bekom-
men sie selten/ da sie Frchte trgen. Barthol. anglic.lib. 17 cap.88.
[They readily grow on stony, dense land / and in cold air/ but warm air cannot do
them much harm. On dense and loamy ground it seldom happens / that they bear
fruit] (transl. VW).
Bartholomaeus Anglicus was a Franciscan friar in the first half of the thirteenth
century, who wrote an encyclopaedia (De proprietatibus rerum), which was used
until the sixteenth century. It was also translated into several vernacular languages.
Exchange between agricultural textbooks and encyclopaedias worked both ways,
from the specialised to the general, and vice versa. Pliny, Albertus Magnus, Konrad
von Megenberg and others have to be viewed as part of the transmission network
of agricultural knowledge.
Coler obviously speaks about two kinds of soil, which he describes with
the same word fest (dense). Coler gives no further detail, so we cannot assess what
his distinction between dense and stony on one hand and dense and loamy on the
other hand really means, but altogether his opinion is contrary to all the previously
cited works, which were in agreement about loose and soft soil as being well suited.
But there is agreement as to loamy conditions not being suitable.
Loudons Encyclopaedia of 1883 has more detail. A table shows soils suitable
for trees. Chestnut trees are considered suitable for topsoils consisting of flinty and
gravelly loam, with a subsoil composed of chalk at four feet with deep gravelly
loam and a few flints, they are also recommended for gravelly and chalky loams
with the identical subsoil, and for sandy gravel and sandy loam, with gravelly
and sandy loam as subsoil.
Consulting the internet to find out what current wisdom has to offer, one
finds: The most suitable soil for Chestnut trees is a sandy loam, with a dry bottom,
but they will grow in any soil, provided the subsoil be dry (http://www.botanical.com/
botanical/mgmh/c/cheswe59.html). Also: Geeignete Bden verfgen ber lockeres,
tiefes, feuchtes, karbonat- und salzarmes Erdreich [suitable lands have loose, deep,
humid soil devoid of carbonate and salt] (http://www.frujorge.com/sorayafruits/
aleman/castanas.htm trans. VW). And finally, Sie vertrgt Schatten, meidet Kalk,
braucht nhrstoffreiche, tiefgrndige Bden, mildes Klima [(The chestnut) tolerates
shade, avoids lime, and needs nutrient-rich and deep soils] (http://www.biozac.de/
biozac/capvil/Cvcastan.htm trans. VW). The very last quote bears more similarity
to the classical authors than Coler or Loudon or the other two internet resources.
What can one conclude from this example?
The history of soil knowledge is indeed full of contradictions, (mis-)inter-
pretations and at times seems circular. The difficulty in describing a phenomenon
like the soil in inter-subjective categories is still discernible today. But this is all
211
Prolegomena to a History of Soil Knowledge in Europe
the more reason to respect the cognitive and linguistic achievements of the first
writers on soils.
Concluding Remarks
Reflected in the title Prolegomena is the notion that this chapter will be a pro-
logue to future studies on the history of soil knowledge. There is not much point
in writing a conclusion to the beginning of something, but I can sum up my initial
arguments. I have attempted to show that knowledge, and in particular the author-
ity granted to knowledge, are concepts which are best understood as interactive. I
have attempted to make clear the amount in numbers and the wide diffusion of the
body of texts, which are part of the interaction of producing knowledge, and have
tried to contextualise the texts within the overall mentality of agricultural societies.
Historiography has not conceptualised texts about soils as one body to be studied
in its entirety so far, so by re-writing I construct a new knowledge system.
I have tried to make clear that the transformation from the perception of
a handful of soil into a text describing it is complicated and by no means easy.
Terminology cannot be taken for granted but must be understood as the result of
a conscious attempt of transformation of sensory perception. Other than handfuls
of soil, texts are a primary agent of interaction in the textual network. I have tried
to make the discontinuities and vagaries of textual transmission visible using the
examples of Mago and of the Geoponika. I have also tried to show the influence
of the requirements of the narrative on the conceptualisation of soil knowledge,
and I have tried, finally, to show processes of differentiation and synthesis of texts
at work.
I hope it has become clear that the sheer amount of available information
prevents any general picture within the limited space of this article. An attempt to
describe the entirety of textual tradition would have to proceed in several (inter-
related) stages. One would need to show the interrelation between the classical
texts and other texts now lost that the authors refer to. Then one would have to
make visible the dissemination of classical knowledge by listing all manuscripts and
their life-histories, their probable place(s) of origin and of use, their travels and the
reworking of their texts, paying attention to what was excluded or included in the
process. It would be necessary to trace all extant translations of classical texts, and
relate them to manuscript traditions, and to investigate the use of classical texts in
several genres of literature such as encyclopaedias and specialised works on subjects
such as agriculture, natural sciences or economics. The matter is complicated by
the fact that use of existing information can be direct or indirect, derived from an
original or only from a later user of it. The later the authors write, the more mate-
rial they (theoretically) have at their disposal, so the question arises why authors
grant authority to a writer they use, or whether they would have had the choice
212
Verena Winiwarter
of using others but did not. Sometimes some, if sketchy, evidence on that can be
unearthed. One would be looking for breaks and interruptions, for misunderstand-
ing and for adaptation, as well as taking into account the question of availability
and abilities of the authors. The examples I gave are certainly a part of this larger
picture whether and, if so, how their interpretation would change within the
sketched larger framework, will have to be left for future study.
Acknowledgements
Thanks (as so often) go to Herwig Weigl, who not only read this text, but also helped with
translations, and thus made it a lot better; to Karl Brunner, who helped locating the Church
Father references; and to Martin Schmid, who made valuable suggestions on how to structure
the piece. Funding by the Austrian Academy of Sciences, APART (Austrian Program on
Advanced Research and Technology) Grant 10916, Precious Dirt Underfoot. A History of
Soil Knowledge enabled the preparation of this chapter.
Sources
Albertus Magnus. De vegetabilibus libri VII, historiae naturalis pars, XVIII, ed. Ernst Meyer
and Karl Jessen. Berlin: G. Reimer, 1867.
Caesarius of Arles. Caesarii Arelatensis opera, I: Sancti Caesarii Arelatensis sermones, ed. G.
Morin, 2nd edn (Corpus Christianorum, Series Latina, Vol. 103). Brepols: Turnhout,
1953.
Cato, agr. = M. Porcius Cato, De agri cultura, Latin/English, ed. William Davis Hooper
and Harrison Boyd Ash, Loeb Classical Library. Cambridge MA: Harvard University
Press 1934.
Col. r.r. = L. Iunius Moderatus Columella, De re rustica, Latin/German, ed. Will Richter et
al. Munich: Tusculum, Vol. 1 1981, Vol. 2 1982, Vol. 3 1983.
Gargilius Martialis (Q. Gargilii Martialis). De hortis, ed. Innocenzo Mazzini, Opuscula
philologica 1. Bologna: 1988.
Heriger of Lobbes. Gesta pontificum Tungrensium et Leodiensium, Monumenta Germaniae
Historia, Scriptores 7, ed. G. Pertz (Hannover: 1846), pp.162238.
Hieronymus (Jerome). Commentariorum in Esaiam libri XII-XVIII. In Esaiam parvula ad-
breviatio, ed. M. Adriaen, G. Morin, (Corpus Christianorum Series Latina Vol. 73A).
Brepols: Turnhoult, 1963.
Konrad von Megenberg. Werke, konomik, ed. Von Sabine Krger. In MGH, Staatsschriften
des spteren Mittelalters, Vol. 3. Stuttgart: 1973.
Pall. Op.agr. = Rutilius Taurus Aemilianus Palladius, Opus Agriculturae, ed. Robert H.
Rodgers. Leipzig: 1975.
Pall. Op.agr. = Rutilius Taurus Aemilianus Palladius, Opus Agriculturae I and II Latin/French,
ed. Martin Ren. Paris: 1976.
213
Prolegomena to a History of Soil Knowledge in Europe
Petrus de Crescentiis. Ruralia Commoda: das Wissen des vollkommenen Landwirts um 1300,
ed. Will Richter, foreword by Reinhilt Richter-Bergmeier, Editiones Heidelbergenses
2527, 30. Heidelberg: C. Winter 19952002.
Plin. n.h. XVII = C. Plinius Secundus [the elder], Naturalis Historiae Liber XVII, Latin/
German, ed. and trans. Roderich Knig and Joachim Hopp. Munich: Artemis &
Winkler/Tusculum, 1994.
Plin. n.h. XVIII = C. Plinius Secundus [the elder], Naturalis Historiae Liber XVIII, Latin/
German, ed. and trans. Roderich Knig, Joachim Hopp and Wolfgang Glckner. Mu-
nich: Artemis & Winkler/Tusculum, 1995.
Rufinus. Rufin dAquile. Les Bndictions des Patriarches, Sources chrtiennes, 140, ed.
Manlio Simonetti. Paris, 1968.
Varro, rust. = M. Terentius Varro, Res Rusticae, Latin/German, ed. and trans. Dieter Flach,
Gesprche ber die Landwirtschaft. Darmstadt: Vol. 1 1996, Vol. 2 1997.
Verg. Georg. = P. Vergilius Maro (Virgil), Georgica, Latin/German, ed. J. and M. Gtte et
al. Munich: Tusculum, 5th edn. 1987.
Vulgata. http://www.vatican.va/archive/bible/nova_vulgata/documents/nova-vulgata_vetus-
testamentum_lt.html (Jeromes translation of the Greek and Hebrew Scriptures into
the common language, Latin, was completed in 405. It was recognised as authoritative
during the Council of Trent (1546) and became the official Bible of the Roman Catholic
Church. The widespread use of the Vulgate is also recognisable in its influence in early
modern Bible translations, such as the Authorised, or King James, Version. The Vulgate
continues to be of scholarly use today in the study of the textual transmission of the
Bible and in the historical study of Christian theology. The Vulgate used is the fourth
edition of the Biblia Sacra iuxta vulgatam versionem. It was originally published in 1969,
and this fourth edition was released in 1994.)
Xenophon. Oikonomiks. Xenophon, konomische Schriften. ed. and transl. Gert Audring,
Schriften und Quellen der Alten Welt. Berlin: Akademie-Verlag, 1992.
References
Barbiero, Daniel. n.d. Tacit Knowledge. In Chris Eliasmith (ed.), Dictionary of the Philosophy
of Mind, http://www.artsci.wustl.edu/~philos/MindDict/tacitknowledge.html
Butzer, Karl W. 1993. The Classical Tradition of Agronomic Science: Perspectives on Caro-
lingian Agriculture and Agronomy. In P.L. Butzer and D. Lohrmann, Science in Western
and Eastern Civilization in Carolingian Times (Basel: Birkhuser), pp.540605.
Butzer, Karl W. 1994. The Islamic Traditions of Agroecology: Crosscultural Experience,
Ideas and Innovations. Ecumene, 1: 749.
Clarke, Roger. 2001. Knowledge. http://www.anu.edu.au/people/Roger.Clarke/SOS/
Know.html
Davidson, Arnold I. 1986. Archaeology, Genealogy, Ethics. In David Couzens Hoy (ed.),
Foucault: A Critical Reader (New York: Basil Blackwell), pp.22133.
Dreyfus, Hubert L. and Paul Rabinow. 1982. Michel Foucault: Beyond Structuralism and
Hermeneutics. Chicago: University of Chicago Press.
214
Verena Winiwarter
Foucault, Michel. 1972. The Archaeology of Knowledge, trans. A.M. Sheridan Smith. London:
Tavistock.
Foucault, Michel. 1980. The History of Sexuality. Vol. 1. New York: Vintage.
Foucault, Michel. 1984. Nietzsche, Genealogy, History. In Paul Rabinow (ed.), The Foucault
Reader (New York: Pantheon), pp. 76100.
Gingrich, Andre and Elke Mader (eds). 2002. Metamorphosen der Natur, Sozialanthro-
pologische Untersuchungen zum Verhltnis von Weltbild und natrlicher Umwelt,
series editors Marianne Nrnberger, Ulrike Davis-Sulikowski and Wolfgang Kraus.
Vienna/Cologne/Weimar: Bhlau.
Glacken, Clarence J. 1967. Traces on the Rhodian Shore. Nature and Culture in Western Thought
from Ancient Times to the End of the Eighteenth Century. Berkeley/Los Angeles/London:
University of California Press.
Gntz, Max. 1897. Handbuch der landwirtschaftlichen Literatur. Mit biographischen Notizen und
38 Bildnissen von Autoren. 2 vols. Leipzig 18971902, reprinted Vaduz: Topos 1977.
Hussain, Athar. 1971. A Brief Resum of the Archaeology of Knowledge. Theoretical
Practice, 3/4: 1047.
International Council of Scientific Unions. 2002. Science, Traditional Knowledge and Sus-
tainable Development, ICSU Series on Science for Sustainable Development No. 4.
Paris: ICSU.
Jacob, Franois. 1972. Die Logik des Lebenden. Von der Urzeugung zum genetischen Code.
Frankfurt: S. Fischer. (English, The Logic of Life: A History of Heredity, trans. Betty E.
Spillmann. Princeton University Press, 1993.)
Kallfelz, Hatto (trans.). 1973. Vita Bennonis II. Episcopi Osnaburgensis auctore Norberto
abbate. In Lebensbeschreibungen einiger Bischfe des 10.12. Jahrhunderts (= Ausgewhlte
Quellen zur deutschen Geschichte des Mittelalters, Freiherr vom Stein-Gedchtnisausgabe
Vol. XXII [Darmstadt: Wissenschaftliche Buchgesellschaft], pp.363441).
Keiser, George R. 1998. Works of Science and Information, A Manual of the Writings in
Middle English 10501500, series editor Albert E, Hartung, XXV. New Haven: The
Connecticut Academy of Arts and Sciences.
Klingshirn, William E. 1994. Caesarius of Arles. The Making of a Christian Community in
Late Antique Gaul. Cambridge: Cambridge University Press.
Le Goff Jacques, Roger Chartier and Jacques Revel (eds). 1990. Die Rckeroberung des his-
torischen Denkens. Grundlagen der neuen Geschichtswissenschaft. Frankfurt: Fischer.
Lefort, Jacques. 2002. The Rural Economy, Seventh-Twelfth Centuries. In Angeliki Laiou
(ed.), The Economic History of Byzantium: From the Seventh through the Fifteenth Century,
Dumbarton Oaks Studies 39. Washington: Dumbarton Oaks Research Library.
Loudon, J[ohn]. C[laudius]. 1883[1839]. An Encyclopaedia of Agriculture: Comprising the
Theory and Practice of the Valuation, Transfer, Laying Out, Improvement, and Manage-
ment of Landed Property, and of the Cultivation and Economy of the Animal and Vegetable
Productions of Agriculture. 8th edn., 1883. London: Longmans, Green, and Co.
Mahaffy, J.P. 1889/90. The Work of Mago on Agriculture. Hermathena, 7: 2935.
Mattey, George J. 2002. Lecture Notes, Lehrers Theory of Knowledge, 2nd edn., Chapter 1,
The Analysis of Knowledge http://www-philosophy.ucdavis.edu/phi102/tkch1.htm
215
Prolegomena to a History of Soil Knowledge in Europe
McDonald, Donald. 1908. Agricultural Writers from Sir Walter of Henley to Arthur Young,
1200-1800. New York: Burt Franklin, reprinted 1968.
McKitterick, Rosamond. 1988. The Carolingians and the Written Word. Cambridge/New
York: Cambridge University Press.
Meyer, Ulrich. 1998. Soziales Handeln im Zeichen des Hauses. Zur konomik in der Sptantike
und im frheren Mittelalter, Verffentlichungen des Max-Planck-Instituts fr Geschichte,
140. Gttingen: Vandenhoeck und Ruprecht.
Rex, Hannelore. 2001. Die lateinische Agrarliteratur von den Anfngen bis zur frhen
Neuzeit. Phil. Diss. Bergische Universitt - Gesamthochschule Wuppertal.
Rheinberger, Hans-Jrg. 2001. Experimentalsysteme und epistemische Dinge Eine Geschichte der
Proteinsynthese im Reagenzglas.Wallstein. (English, Toward a History of Epistemic Things:
Synthesizing Proteins in the Test Tube. Stanford University Press, 1997).
Rice, John S. 1992. Discursive Formation, Life Stories, and the Emergence of Co-De-
pendency: Power/Knowledge and the Search for Identity. Sociological Quarterly,
33(3):33764.
Rodgers, Robert. 2002. Kepopoia: Garden Making and Garden Culture in the Geoponica. In
Anthony Littlewood, Henry Maguire and Joachim Wolschke-Bulmahn (eds), Byzantine
Garden Culture. Washington: Dumbarton Oaks Research Library.
Verhulst, Adriaan E. 1989. Agrarian Revolutions: Myth or Reality? Sartoniana 2: 715.
Websters New Twentieth Century Dictionary. 1983. Simon and Schuster: New York.
White, Kenneth Douglas. 1970. Roman Farming, Aspects of Greek and Roman Life. Lon-
don: Thames and Hudson.
Winiwarter, Verena. 1992. Landwirtschaftliche Kalender im frhen Mittelalter. berlegun-
gen zum Fortleben antiken Bildungsgutes und zu seinen Funktionszusammenhngen.
Medium Aevum Quotidianum, 27: 3355.
Winiwarter, Verena. 1999. Bden in Agrargesellschaften: Wahrnehmung, Behandlung
und Theorie von Cato bis Palladius. In Rolf Peter Sieferle and Helga Breuninger (eds),
Natur-Bilder. Wahrnehmungen von Natur und Umwelt in der Geschichte (Frankfurt:
Campus), pp.181221.
Winiwarter, Verena. 2000. Soils in Ancient Roman Agriculture: Analytical Approaches to
Invisible Properties. In H. Novotny and M. Weiss (eds.), Shifting Boundaries of the Real:
Making the Invisible Visible (Zurich: vdf Hochschulverlag), pp.13756.
Winiwarter, Verena. 2001. Nachhaltigkeit. Zur Paradoxie von Theorie und Praxis. In
Grg, Bernhard, Innovation und Verantwortung. Werte, Wege und Aufgaben fr das 21.
Jahrhundert (Vienna: Holzhausen), pp.20519.
Worster, Donald. 1988. Dust Bowl. Drre und Winderosion im amerikanischen Sd-
westen. In Rolf Peter Sieferle, Fortschritte der Naturzerstrung (Frankfurt: Suhrkamp),
pp.11857.
Ziman, John. 1976. The Force of Knowledge. The Scientific Dimension of Society. Cambridge:
Cambridge University Press.
216
Robert S. Shiel
Robert S. Shiel
Introduction
Europe is fortunate in that, with the exception of the far north and highland regions,
there is sufficient warmth and moisture to grow wheat, barley, oats or rye with good
reliability provided the soil has an adequate supply of nutrients and is not exces-
sively wet. Along the Atlantic fringe drying and ripening of the crop may be made
difficult by the wet climate. In the southern tip of Spain, Greece and Italy drought
is common but there is usually sufficient winter moisture for barley or millet to
be grown. The range of fruit, vegetable and cereal crops and animals that can be
produced in any location varies but, with the above exceptions, there is always a
range of potential food products available. Nearly half the soil is cambisols, which
are inherently fertile, but one quarter is podzols and luvisols, which contain fewer
nutrients and may be excessively acidic. Under the widespread natural deciduous
woodland, these soils changed fertility slowly and accumulated a large reserve of
nitrogen; soils in those areas with coniferous woodland contained fewer nutrients.
After clearance of the woodland, nutrients were lost rapidly by increased leaching,
soil erosion and offtake in crops; crop yield under a monoculture decreased to a
low level within one human generation and the soils would not then yield adequate
crops to repay the effort of cultivation. The soil nutrient status recovers more
slowly, if the natural vegetation is allowed to re-establish, than it decreases during
exploitation and other land will have to be used during this recovery phase. Such
shifting cultivation depends on having a large excess of cultivable land on which
recovery proceeds, and if this is not available then techniques to maintain fertility
are required. These include the application of wood ash, the use of animal manures
and the recycling of human waste. It may also be possible to import material such
as marl to raise the pH rapidly. Several of these methods maintain fertility on the
manured area but result in slow fertility loss on the large areas of grassland and
217
Nutrient Flows in Pre-Modern Agriculture in Europe
woodland exploited for manure or ash; fallowing increases crop yield but leads to
a more rapid decrease in soil organic matter and nitrogen. The intentional use of
legumes and deep rooting plants, respectively, introduce external nitrogen and make
other nutrients more available within the root zone of the plant. The sophisticated
rotations of the eighteenth century maintained fertility much better than had the
earlier rotations but still depended on large amounts of animal manure or ash from
outside the rotated area if the yields were to be commercially attractive and sustain-
able. The modern concept of organic farming eschews inputs of water soluble and
most exotic materials and is able to maintain yields at a substantial proportion of
those obtained using conventional farming methods. It does however make use of
external natural fertilising materials such as rock phosphate and limestone, which
were not widely available or used in the distant past, but remain endemically short
of plant nutrients. The tradition of recycling urban waste night soil, which could
replace much of the nutrient loss, is less acceptable today on aesthetic grounds and
much of that collected is now contaminated with toxic industrial wastes.
In this paper I will examine the original pattern and properties of soils in
Europe before the initiation of widespread land management as we recognise it today
and will then look at the natural and human-induced alterations to those soils. This
will involve looking at the sustainability of earlier land use systems and will be based,
where possible, on sound scientific evidence from field research and experiments.
The soils of Europe have for long been modified by the human population. In some
other parts of the world soil modification, on other than a hunter-gatherer scale,
has been restricted to the last two hundred years. In these places there are experi-
ments which, arguably, allow us to reflect on the changes that occur when virgin
sites are first used. In Europe, there is also an established literature on agricultural
management of land dating back to Hesiod (West 1988) in the eighth century
bc (See Winiwarter in this volume); the role of these management techniques in
mitigating deterioration and even in ameliorating the soil will be examined. Ero-
sion and its corollary, deposition, have influenced the nutrient supply throughout
the period but, as they are fundamentally dependent on the vegetation, which has
been heavily altered by human action, their effect will be considered together with
the management of the land rather than as a separate topic.
barley, which is the most drought and salt tolerant, to oats and rye, which tolerate
moister, more acidic environments with short growing seasons (Leonard and Martin
1963). On the wetter western fringe, ripening and drying cereals can be difficult
and here dairy products and the more recent introduction, potato, are preferred.
The upper and northern boundaries between the areas that are more or less suitable
for particular crops have varied with temperature in the last millennium with
the variation ranging from the Little Optimum through the Little Ice Age (Parry
1978) into the modern amelioration. The pattern of rainfall varies from a summer
maximum between about 45 to 56 N to a very distinct late autumn maximum in
the south (Table 8.1). The furthest north locations also have an autumn maximum
but not the summer minimum of the southern sites. All have on average more
than the total needed for wheat growth and as, in the driest areas, the rain falls
predominantly in the cool months when wheat is grown, there will be relatively few
sites at which wheat fails completely on a regular basis. The effect of the moisture
balance on crop growth has, like the temperature regime, varied considerably over
the last millennium and this variation is most likely, as with the temperature, to
affect growth near the extremes.
The parent materials from which soils have formed are very variable. The
continent has been affected by a number of orogenic episodes, the most recent
being the Alpine, all of which have led to intense denudation of the uplands and
deposition of alluvium in valleys on mountain fringes. In the Mesozoic however
much of the continent was flooded and, being on the edge of the Tethys Ocean,
is underlain by younger marine rocks. The main areas of ancient metamorphosed
rocks lie along the western edge. These rocks weather and release minerals for
plant growth slowly, whereas the younger rocks provide a more on-going supply.
Although, in earlier time, the continent lay at low latitudes and suffered intense
chemical weathering, its more recent northern location has led to much of the
weathered rock being eroded or capped with recent alluvial or glacial deposits rich
in plant nutrients. The glaciation affected the northwest and mountains directly
(Flint 1971), where it left areas of glacial till, and other regions were indirectly
affected by the deposition of large areas of loess and outwash deposits. The least
fertile of these deposits are the fluvio-glacial sands, which contain few weatherable
minerals. In the areas that were not glaciated there has been severe episodic erosion
of soil, especially in the drier areas where vegetation cover is poor. Although this
erosion did not always move soil far, it frequently reduced the total area of soil
without producing a compensatory increase in soil quality in the areas on which
eroded material was deposited (van Andel 1987). A very active period of erosion
occurred at the end of the glacial when plant communities were destabilised and
there was an increase in intense rainfall events.
219
Nutrient Flows in Pre-Modern Agriculture in Europe
Table 8.1. Monthly and annual rainfall averages (mm) for European cities between
37 and 60 latitude. The three wettest months are marked
Latitude 37 41 42 43 45 48 48 52 56 59 60
Stockholm
Florence
Helsinki
Moscow
Madrid
Toulon
Vienna
Athens
Berlin
Rome
Paris
January 48 45.1 80 48.7 64.1 38 45.8 43.2 34.4 31.4 46
February 41 43.2 70.9 39.5 61.5 37.2 38.9 37.6 29 25.3 36.6
March 41.2 36.8 68.6 44.5 69.4 44.2 41 37.8 32.7 26.3 34.9
April 23.4 45.4 66.8 46.2 70.5 51.3 44.4 41.1 38.2 28.6 37.4
May 17.9 39.7 51.5 46.1 73.3 68 55.7 49.4 51 33.9 42.1
June 7.4 25.2 34.1 26 56.4 70.9 56.6 64.1 65.6 44 45.9
July 5 9.4 16.3 14.6 34.2 74.6 57 71.1 81.5 64.4 61.5
August 7.6 10 24.4 24.3 46.9 68 55.2 62.1 71.8 66.1 74.7
September 9.8 29.3 69.2 63.4 83.4 47.5 52.8 44.1 57.7 48.7 66.5
October 53 46.4 113.3 93.6 99.1 49.1 56.5 44.3 50.4 50.6 68.5
November 55.3 63.9 110.7 76.3 103.4 47.7 53.7 45.5 44.1 44 65.7
December 61.8 47 97.1 58.6 79.4 45.8 49.3 47.9 42.4 39 55
Total 371.4 438.9 802.9 582 842.1 642.9 607.4 588.7 600.6 502.8 635.4
Source: http://www.worldclimate.com/
ing. Lithosols (5per cent) occur on steep slopes where there is much rock exposed
and the environment is usually only suitable for low-intensity grazing. Rendzinas
(5per cent) are found on the extensive chalk and limestone outcrops where these
are not covered by non-calcareous deposits. The soils are often thin and can be
droughty if the underlying rock is impenetrable to roots. They provide grazing and
cereal production but, in addition to drought, crops can suffer from micronutrient
deficiencies. Including the Eastern European countries would increase the small area
of phaeozems chernozems which give good cereal growth and are fertile, but
otherwise would have little effect on the overall balance of soils. Locally, the resource
can deviate a long way from these proportions. Thus whole modern countries (eg
Finland) may be dominated by a small, and not always agriculturally attractive,
subset of the soils found across Europe. The natural vegetation over most of the area
would be forest (Eyre 1968); deciduous except in the north and at high elevations,
where conifers are more common, and in parts of the Mediterranean basin, where
in addition to coniferous species there are evergreen broadleaf woodlands. Tundra
is only found in the far north and at high elevation. Natural grassland is restricted
and mostly lies to the east of the Carpathians.
The evolution of the soils has of course continued under both natural proc-
esses and under the modifications to vegetation and soil induced by human activity.
After the initial postglacial increase in soil organic matter and profile development
in the areas more affected by the climatic changes, one of the main changes influ-
encing soil fertility has been the on-going leaching of nutrients from the topsoil.
The amount of leaching that occurs depends partly on the balance between rainfall
and evaporation, and also on the vegetation and soil physical properties. It should
be noted from Table 8.1 above that in average years at all sites although there will
be an excess of rainfall over evaporation, usually in the winter, the amount and
frequency of that excess will vary strongly from year to year and place to place.
In general, the effect on the soils fertility will be proportionally greatest on soils
with a low natural nutrient content in areas with a lower temperature and a larger
winter rainfall. In addition, the amount of leaching will be greater if there is short
vegetation, such as grassland, because it intercepts less of the rainfall before it reaches
the soil (Lockwood and Sellers 1983). Woodlands therefore not only reduce the
amount of leaching but tend to have a greater ability to recycle nutrients from
depth to the surface. Life is not however simple and some trees oak and pine for
example produce polyphenolic compounds which lead to surface acidification
of the soil and even to podzol formation. Thus, under woodlands a spectrum of
soils from brown earth to podzol can be found. Once the natural vegetation cover
was complete there would have been little opportunity for soil erosion, and soil
profile development would have continued unhindered. Evidence in major river
valleys indicates a fining of the sediments upwards which supports the view that the
landscape became relatively stable at an early date. Thus once the climax vegetation
221
Nutrient Flows in Pre-Modern Agriculture in Europe
was established the rate of soil change slowed. Soils tended to have a large organic
matter content with a large store of nitrogen, a good soil structure of vertical pores
down which water could drain and roots grow and, with active recycling of nutri-
ents from depth to the surface, the loss of nutrients was smaller than under shorter
vegetation. Nutrients consumed by animals tended to be recycled locally. The effect
of the postglacial climatic amelioration and the soil and vegetation succession was
therefore to provide conditions that were attractive for the growth of a wide range
of plants, which could provide food for a variety of animals. The areas that were
too cold or where the soils were too wet or heavy were the main locations that
would have been unattractive to early farmers.
matter contents and the most rapid rate of organic matter degradation. They are also
the part of Europe in which there is currently a small risk (~1:20) (Arnon 1972) of
summer drought severe enough to prevent wheat growth. Floodplain soils, though
the most fertile in these and other areas, have a relatively large risk of damaging
spring or autumn floods, which would be exacerbated by clearance of vegetation
from surrounding slopes (Shiel 1997).
The clearance of land by fire can also expose the soil to water erosion in the
period between burning and the regrowth of plant cover. This problem and the
risk of loss of the nutrient-rich ash by leaching or wind erosion (Sillitoe and Shiel
1999) will be worst in dry areas with episodic intense rainfall. Plato lamented,
as a folk memory in the fourth century bc (Lee 1971), the loss of soil from the
hillslopes in Attica, in a region with a mean annual rainfall of about 450mm but
where there is currently a one year in 10 risk of a month with 200mm rain and a
similar risk of three consecutive months having in total 400mm rain (Shiel and
Stewart 2003). In the areas which do not have a carbonate-rich subsoil, removal
of the woodland will lead to increased leaching of bases so that the soil becomes
progressively impoverished in plant nutrients and also becomes acidic. It seems
therefore that although the properties of the soils over a large part of Europe
ameliorated greatly in the early postglacial, even the low intensity use of land for
grazing and opportunist collecting of fruits, associated with the removal of com-
petitor species may have started a process of soil deterioration, with increased soil
wetness, loss of organic matter, leaching of nutrients and reduction of soil structure
and moisture retention.
The animals void over three quarters of the plant nutrients ingested, though
some of the nitrogen is lost in forms that are not available to plants. The nutrients
are not, however, always returned where they were consumed and hence the animals
may alter the fertility of the soil with some areas being enriched while others are
depleted. The most important other requirement of the shallow rooting species is
a continuing supply of water, particularly during the growing season. Variation in
this has been shown to have a major effect on forage yield. Without an adequate
water supply, growth stops or at best slows during periods when otherwise tem-
perature and light are excellent for growth. During summer and early autumn
in the drier areas of south and east Europe growth stops for all but the deepest
rooting species or where there is a high water table at this time stock will eat any
remaining standing vegetation or will browse trees and shrubs. In many areas it
has become normal for stock to be moved during the summer to wet floodplains
or to mountains where there is a better supply of water transhumance. Over all
Europe bar the very western fringe and extreme south it is too cold during at least
part of the winter and forage growth ceases. The length of this season determines
the extent to which the farmer must make alternative arrangements for animal
feed, usually by conserving some of the spring or autumn excess as hay. Farmers
may even decide to feed some hay during the summer in areas with a prolonged
drought. The amount of nutrients removed from the soil in conserved forages is
much larger than that removed by grazing animals; it incidentally is also greater
than the amount removed in the majority of arable crops, and so could result in
rapid depletion of soil fertility. Conversely, the animals will leave areas where they
are fed and housed in winter enriched with nutrients. If manures can be collected,
these are of particular benefit in maintaining fertility of cultivated soils, but most
are not applied to the soil from which they originate. We can therefore conceive
of zones of enrichment and depletion; the enrichment is often close to the habita-
tion area with the impoverishment occurring around the fringe (Shiel 1991). In
seasonal grazing areas which are distant from the permanent cultivated area there
is commonly little nutrient relocation, other than that which occurs locally into
areas where animals are penned at night or during the heat of the day.
If the first spring growth is cut for forage on part of the land then, if un-
manured, it will probably yield about 2t/ha of dry matter, as it has at the long
term British experiments (Hall 1905; Shiel 2002) traditionally as hay. The ad-
dition of minerals from say wood ash will increase the hay yield to about 4t/ha
dry matter. Although manure would increase hay growth further, to over 5t/ha,
it is doubtful if this would be the most advantageous use of the manure; it would
enable more people to be fed if applied to cereals. The spring growth of grass will
be less affected by local rainfall than is the later growth, as for all but the western
edge of Europe there is a large water deficit which becomes severe from the end of
May (Table 8.1). Where water is available during the summer, growth continues
224
Robert S. Shiel
and will give about another 2t/ha but in the drier areas there will be less than an
extra tonne. If the conserved forage is fed to animals which are kept indoors then
about twice as much fresh weight of manure will result compared to the weight of
the dry material fed (Shiel 1991). Growth of grass on land which is grazed in situ
and is not cut regularly for hay will tend to be substantially larger than these figures
because of the nutrients returned in the urine and manure; there will of course be a
correspondingly smaller amount of manure available to spread on arable land.
Table 8.2. The energy and protein content of four major temperate cereals and their
yields according to sources in Norfolk and Hampshire, England during the last
millennium.
Energy and protein in grain*
energy protein energy protein
kcal/100g % wheat =1
wheat 348.2 14.1 1 1
barley 334.2 11.8 0.96 0.84
rye 366.8 16.2 1.05 1.15
oats 319.8 13.9 0.92 0.99
two people. This however ignores losses due to vermin, waste and rotting as well
as that for seed; equally it ignores other sources of food. As these values are from
inventories they may of course be tactical underestimates of the crops but they are
seen later (Table 8.3) to be comparable, in the case of wheat, with yields from the
unmanured plots at Broadbalk in the nineteenth century.
high in pH and several nutrients (Sillitoe and Shiel 1999; Bear 1931); the vigorous
growth of plants near fires, on burnt land and where ash had been fortuitously scat-
tered would have been as obvious to the street wise cultivators of the past as was
the observation that plants grew better near manure. Use of these materials would
allow a monoculture to persist for longer but it would eventually fail because the
soil became difficult to manage, weeds became rife or pests and diseases destroyed
the crop. The land would then be abandoned, perhaps to be reused at a later date
shifting cultivation. Additional sophistication involves not growing the same
species every year on the same land or in not using the land at all for one year fal-
lowing. This reduces the pest, weed and disease problems and may also affect the
rate at which nutrient reserves in the soil are depleted. Certainly such a rotation of
crops was not news to Virgil (Wilkinson 1978); his rotations describe the benefits
from wild forage legumes such as vetches and Columella (Col.r.r. II, 7, ed. Ash
1941) indicates that a wide range of these were intentionally sown for forage for
animals. Surprisingly, however, this knowledge appears to have been subsequently
mislaid, and in England it was not until the eighteenth century ad that farmers
were encouraged by the popularisers of the new Agricultural Revolution to sow
particular forage species within a rotation, rather than sowing the mixture of grass
and other seeds swept up from the floor of the hay loft (Moore 1944).
The use of non-waste soil ameliorants also originates from an early date;
Pliny the Elder reports the use by the native inhabitants of Britain and Gaul of
marl to improve soil (raise the pH) (see Chapter 7). The benefits of irrigation were
also understood, though the role of the suspended silt in floodwaters may have
been confused with the impact of the water itself. Silt from rivers was also used
intentionally until recently in warping (Ferro 1949) and may have been exploited
unwittingly in the case of the management of water meadows. These remained the
main external ameliorants until the use of ground bones in the eighteenth century.
Other external materials include marine resources seaweed, shell sand and surplus
fish, but the use of these seems to have been limited geographically.
How much of the classical knowledge survived the Dark Ages after the fall
of the Western Roman Empire is as yet unclear, although the Venerable Bede, writ-
ing in Northumbria in the seventh century ad, was certainly quoting widely from
Roman sources. As his writings penetrated through Europe fairly efficiently and the
Germanic tribes who settled the north and west of Europe had a well established
farming system using the same range of crops and animals, diet and land manage-
ment may have continued with less change than might be assumed. The origin
of the mediaeval 3-field rotational system in this period is uncertain but it was in
place with the fully fledged feudal system at the beginning of the second millen-
nium ad. The knowledge necessary to modify the 3-field system into the rotations
which remain in use today was in place from the middle of the second millennium
ad but it was the eighteenth century before the intentional use of root crops (e.g.
227
Nutrient Flows in Pre-Modern Agriculture in Europe
turnips) and forage legumes became widespread; the reasons for the delay seem to
have been as much social as technological. Even when the sophisticated rotations
were introduced, fallowing, which was replaced by the root crops, did not com-
pletely disappear. It must be noted that although these systems were used in such
a way that fertility appeared to be managed satisfactorily, the only external sources
of ameliorant involved the use of marl and biological nitrogen fixation. All of the
other methods of maintaining fertility of the arable land operated by the invidious
procedure of robbing Peter to pay Paul! Thus animal manure spread on arable land
decreased the nutrients in the surrounding grasslands and woodlands, wood ash use
similarly debilitated the forests from which the wood was taken, river silt settling on
irrigated land degraded the soils from whence it was eroded and even human wastes
were rarely applied to the soils from which the nutrients had been taken. The effect
of these practices was that fertility diversified within the landscape with those areas
which received a net surplus of nutrients being ameliorated and the others being
degraded; this can have a long-term effect on future fertility. Russell (1915) notes
that the phosphate in manures applied to fertilise fields in Egypt in the pharaonic
period could still be measured. The extent to which nutrient leaching and erosion
of soil occurs also varies with crop grown and system of production. Compared
with grassland, all arable crops leave the soil bare at some season, facilitating ero-
sion, and even the longest-season cereal wheat evaporates a lot less water than
grassland, increasing the leaching of nutrients (Lockwood and Sellers 1983). A
rotation of crops can reduce the risk of both leaching and soil erosion by ensuring
that the soil has a cover of vegetation for more of the year than is possible with a
monoculture. The worst option of all is fallow, which leaves the soil intentionally
exposed and where it is subject to the effects of intense rainfall and strong winds,
and where the amount of water available to leach out nutrients is much larger than
when a crop is grown.
weight of the wood. As the wood consumed annually might be in the region of
1t/person in a warm area to 3t/person in a cold area (Lefevre et al. 1997), a group
of 10 people burning 1030 tonnes of wood would only produce 200600kg
ash containing proportionately more calcium and potassium than manure but no
nitrogen and less phosphorus; ash from five tonnes of wood would give a compa-
rable amount of mineral elements to one tonne of manure. As the manure applica-
tion on the experiments described later (Table 8.3) was 35t/ha the ash from 10
people would at best only manure 0.16ha. In contrast, the manure from animals
housed for half the year and fed conserved feed from 2ha at the average yield for
unmanured grass would produce 8 tonnes manure, and more if night folding were
used sufficient for 0.23ha. The development of urban societies created a traffic
in nutrients; the night soil they produced could be used to enrich peri-urban land
but transport difficulty and expense restricted the distance they could be moved
(Anderson, 1794). People however produce relatively little manure in comparison
to ruminant animals, so from urban societies the main source of manure was their
domestic animals and wastes from food preparation and the ash from cooking
and heating. Some at least of the human waste is likely to have been informally
recycled onto garden plants.
Table 8.3. Yield of Broadbalk continuous wheat plots between 1852 and 1901 and
the percentage of the mean yield in each decade for treatments receiving nothing,
minerals and farmyard manure (FYM) (Hall 1905).
1852 1862 1872 1882 1892 Mean
Treatment 1861 1871 1881 1891 1901 kg/ha
FYM 118.3 126.5 128.5 140.9 154.6 2427
Untreated 55.0 48.9 46.6 46.5 48.5 897
Minerals* 63.7 52.3 54.2 50.9 58.4 1018
*Potassium, Magnesium and Phosphorus equivalent to the nutrients in ashes
balanced combination of nutrients maintains the yield but at a level which depends
on that in least supply, especially if the nutrient in question is N; this is a classic
example of Liebigs (1843) Law of the Minimum. Judging by the yield from the
plot receiving no nutrients the soil fertility at Broadbalk for a range of nutrients
was poor when the experiment began; it had been in continuous arable use before
the experiment began in 1843 and had a low soil organic matter content. Where
the supply of added nutrients is large then the yield increases, possibly due to ac-
cumulation of surpluses in the soil and to the increased return of organic matter
from the larger crop residues. If the mixture of nutrients is imbalanced then yield
trends downwards, the trend rate depending on the magnitude of the store in the
soil of the inadequately provided nutrients. The nearby Hoos Field continuous barley
experiment showed a much greater decrease in yield with time for the unmanured
and mineral treatments than was found at Broadbalk; the yield at the outset was
higher than at Broadbalk but decreased much more rapidly reaching 600kg/ha
within 50 years. The change in yield with time on the differently fertilised barley
plots is very similar to that for the wheat; large balanced inputs of nutrients from
manure maintained yield at a high level throughout the period.
The yield over the first 50 years on the untreated wheat plots at Broadbalk
decreased to just below 900kg/ha and remained about there. This compares favour-
ably with the yields of wheat on the Sanborn untreated plots in Missouri (Brown
1994) which, even averaged over the first 30 years of the experiment, were only
600kg/ha. This figure is similar to that at the Australian Mediterranean climate
zone wheat experiments the yield at Urrbrae in the second decade was 62per
cent of that in the first (Grace and Oades 1994). It has been shown here that
continuous cultivation of a single crop cannot continue unless there is a large area
of land available for manure production and even then it is difficult. If the grass
was unfertilised and all of the manure from hay on half the grassland was collected
then at Broadbalk 17.5ha of grass would be needed to produce sufficient manure
to fertilise 1ha of wheat.
230
Robert S. Shiel
The classical Rothamsted experiments began on fields that have for long
been in use for wheat growing so it is difficult to measure changes in soil nutrient
content. Where the leaching of nutrients was measured the amounts were very
much larger from the plots that were being fertilised (Hall 1905). At Sanborn field
(Brown 1994) there were no original measurements made on the soil and hence
considerable care must be taken in comparing the current properties of the soils
receiving different treatments. It seems that the topsoil thinned considerably and
the organic matter has decreased, especially on the unmanured plots. This loss of
soil on cultivated land is confirmed by many field experiments. For example, over
14 years at a nearby experimental station in Missouri, permanent grass lost 760kg
soil/ha while under continuous wheat the loss was 30 times greater and rainfall
runoff increased from 12per cent to 29.4per cent (Miller and Krusekopf 1932).
This would tend to make the soil droughtier and reduce crop growth. Cultivation
appears to have multiplicative effect on loss of nutrients by removing them physi-
cally in the crop, by increasing leaching of water soluble nutrients, by reducing
the soils ability to retain water and to transport it to depth and by weakening the
structure, as well as by exposing the surface to raindrop impact. Catchment studies
today seem to be more attractive than the chore of the traditional field experiment
and they tend to confirm all that has been said about the risk of soil loss and the
high content of nutrients in the eroded material (Wicherek 1993).
Wind erosion has not been mentioned as it only becomes a problem when
large areas of bare soil are exposed. Where there is a mixture of land uses then soil
eroded from a bare area may be retained in nearby vegetated areas this idea is
exploited today as strip cropping (Gustafson 1941). Although the bare soil may
be degraded the soil remains in the catchment and offsets part of the effect of the
loss from the eroded area. Tillage does however also cause soil movement on a
smaller scale than in classical erosion, where soil is redeposited several kilometres
away. On slopes there is a tendency for loosened soil to move downhill under
gravity resulting in a creep whereby the slopes have a thinner soil than level areas
or footslopes. There is often also localised erosion which contributes to this same
process; over a period of centuries the differences in surface level at either side of
a fence line can be over one metre (Shiel and Chapman 1988). All of these result
in the redistribution of the most fertile fractions of the soil.
1905). Measurements made on the timing and composition of the drainage water
confirm a peak in nitrate content in the water in November and December which
corresponds closely with the months in which the drains flow most commonly. The
result is a substantial peak of nitrate loss from the soil in the final months of the
year which is greatest in wet years and which can restrict future growth if the added
fertiliser is inadequate. In dry areas in contrast spring wheat yield increases linearly
with growing season rain and soil-stored water up to around 400mm (McConkey
et al. 1996). Barley at Rothamsted also yielded slightly more in bushel terms in dry
years but as the bushel weights were smaller there was no clear benefit except on
the minerals (ash) plot where increases of up to 50per cent were obtained (Hall
1905). The benefit on these plots of the dry year is that there is less leaching of the
nitrogen mineralised from the soil over the winter and spring. This is confirmed by a
more detailed analysis which showed that the benefits were largest when the period
March to May was drier than average but June and July were wet. Although the
most apparent effect to the cultivator is on the crop yield, the loss of nutrients by
leaching means that less are recycled and the long-term impact on the soils fertility
is clear. Low yields will also tend to encourage an expansion of cultivated area to
maintain total output; this will encourage even more leaching of nutrients and will
reduce the area of uncultivated land from which manure or ash can be obtained
(Shiel 1991). The driest areas will suffer the least loss of nutrients by leaching, but
may be even more susceptible to the more rapid loss of nutrients by erosion!
and barley lower protein content than wheat (Table 8.2). In terms of total yield
there are large variations in estimated yield between the crops and it may be that
the difference in total output of food value per hectare is relatively small.
For other soil types, changes in soil properties are smaller though organic
matter can decrease rapidly under cultivation (Jenkinson 1988), restricting the
amount of nitrogen available and the ability of the soil to retain water and nutrients,
and erosion can occur. It would therefore be facile to imagine that the soil had
not been altered before neolithic farming began. Archaeological evidence indicates
that though sites may be in use at several periods there may be times when they are
abandoned, often for several hundred years, so even the neolithic cultivator may
have shifted his location if soil fertility decreased. Apart from the narrowing of the
C:N ratio and the short-term benefit from burning even the shifting cultivator is
likely to have left soil more acidic, nutrient depleted, more difficult to cultivate and
more droughty, due to the reduced organic matter content, and possibly wetter and
less well structured than they had been under natural conditions. Shifting cultiva-
tion can be seen as a transition to rotation but by creating a diverse landscape and
breaking up the area of cultivated land and, during the recovery phase, increasing
plant cover it acts to minimise wind erosion and reduce the severity of and localise
water erosion. The extent to which soil recovery occurs during the uncultivated
recovery phase will depend on its length and on the properties of the remaining
soil and underlying parent material. Thus on loess soils, provided some loess re-
mains, erosion has a small effect on future output while if the underlying material
is hard rock or unweathered till then the recovery of fertility will be very slow. At
Rothamsted on a weathered paleosol the organic matter has more than doubled in
81 years since the arable land was abandoned to natural woodland regeneration;
when old grassland was ploughed out however the organic matter content fell to
half in 18 years with a bare fallow. Even with a 6-course rotation including grass,
the organic matter decreased by 30per cent in 26 years (Jenkinson 1988).
on sites which were less easy to manage than the arable land. As more of the virgin
land is exploited there comes a point when land that has been abandoned already
and which has recovered becomes attractive for cultivation, as the other land is
becoming less fertile. This can be viewed as being equivalent to a long rotation; as
it may be physically easier to recultivate formerly cultivated land then this system
has its attraction and is used today in Papua New Guinea (Sillitoe and Shiel 1999).
The alternative to shifting the area cultivated is to alter its use temporarily. The
simplest form of such a rotation is the introduction of a fallow during which no
crop is grown on the land. The next system is to combine this with a change in
crop, although it is feasible to alternate the crops without a fallow. In the absence
of a fallow, even if there is no nutrient advantage to the crop from being grown
in rotation there may be a difference in susceptibility to pests, weeds and diseases
between the crops which results in more of the available nutrients being harvested
in the crops. Finally, one reaches the more sophisticated rotations which include
botanically diverse plants including legumes and the use of the cultivated land
temporarily for animal grazing. Rotation in the Mediterranean seems to have been
in place by early in the first millennium bc and to have continued in this form for
some 2,500 years at least.
Fallows
Because of the problems with pests, weeds and diseases it has not always been
possible to grow the crop on all of the plots in every year in the monocultural
research station experiments even under the high standards of management and
with the tools and technology available. The result of this is that there have been
unintended fallows or changes of cropping; one can be certain that this would have
happened in the past, and the benefits to the crop of growing after a fallow would
have been noted by farmers. When the yield from the Broadbalk wheat grown as
a monoculture is compared with the adjacent Hoos Field plots, where the wheat
is grown after fallow without manure, there is an increase over the 47-year period
from 900 to 1210kg/ha grain and a comparable increase in straw yield due to the
use of the fallow but there is only one crop every alternate year (Hall 1905). If the
amount sown was about 200kg/ha then the net yield on a monoculture is about
650kg/ha and with fallow it is 480kg/ha, after allowing for the uncropped area.
Clearly there have to be other benefits than the larger crop to make fallowing an
attractive proposition. These advantages are that work on the fallow is less critical
than on the crop, the crop is thicker after the fallow and therefore weeds are less
of a problem and, most importantly, the gap allows for weed control of perennials
and also breaks the cycle of pests and diseases. Most of this benefit due to fallow-
ing occurs in the drier years in the 16 years with below average rainfall between
1872 and 1902 the yield increase due to fallow was 51.5per cent while in the wet
years it was only 7.9per cent. In Australia, fallow reduced the rate of yield decrease
234
Robert S. Shiel
with time in the Mediterranean zone in the second decade yield was 94per cent
of the first and in the third 66per cent of the first compared with 60 and 56 per
cent respectively in the monoculture (Grace and Oades 1994). Here the benefits
of fallowing were very much greater than in England, and even allowing for only
half the area being cropped the fallowed plots yielded 12per cent more than the
continuous monoculture wheat. It seems that in the fallow year, as in the temper-
ate region, organic matter breaks down and releases nitrogen which benefits the
following crop. The long-term sustainability of this system is, however, suspect
due to rapid rates of organic matter loss. In the short term (~100 years) fallowing
will increase yields on the cropped area at the expense of cultivating twice the
area cropped but after a century of use both systems will result in a very low soil
organic matter content and the fallowed soil may be even less fertile than the an-
nually cropped land! For a fallow to be effective the next crop must be established
before there is a surplus of water to leach out the accumulated available nitrogen.
There can also be problems as fallow soils are exposed for a considerable time to
wind and water erosion.
In very dry areas fallow also gives an advantage to growth in alternate years
by conserving water in the soil. However, the net effect is the same; in wet years
there is little advantage of fallow because both systems have plenty of water and
in dry years the extra water, and/or nitrogen stored in the soil, benefits the plants
grown on the plots which were fallowed.
All-Arable Rotations
The separation of land used for crops and stock seems to have been widespread,
other than for the temporary use of stock to clean up the waste in fields after
harvest. Many of the rotation experiments carried out include such a compart-
mentalisation even though manure may be applied indicating that stock were
held elsewhere. The simplest of these rotations is the Three field system with the
rotation of wheat, barley, fallow. The closest experiments to these are the Australian
rotations of wheat, fallow, oats at Dooen and Urrbrae in the Mediterranean zone
(Grace and Oades 1994). At Urrbrae the wheat and oat yields on the rotation were
10per cent better than on the wheatfallow rotation; this is excellent as the ratio of
cropped land to unproductive fallow is larger. Allowing for the area uncultivated,
the grain yield averaged over 60 years was 47per cent greater on the rotation than
on the wheatfallow and was 64per cent greater than on the wheat monoculture.
Although the traditional strip farming approach used for such rotations reduced
the risk of wind erosion the presence of a fallow and of long periods between crops
being sown and harvested meant that all of the processes which led to the loss of
organic matter and nutrients from the soil, and of the soil itself by erosion, had
their opportunity to work. These soil losses may explain the lack of increase in
235
Nutrient Flows in Pre-Modern Agriculture in Europe
crop yield during the Middle Ages. Seen in this light, episodes such as the Black
Death may have had the effect of reducing pressure on the soil and giving it the
opportunity to recover whereas during periods of population growth the pressure to
increase total output would have decreased the area of local grassland to the point
that insufficient manure was available to adequately fertilise the arable land. The
reduced crop yield has a vicious cycle effect by exposing the soil to greater leaching,
erosion and organic matter reduction while forcing the farmers to further increase
the area of land ploughed to meet market demands.
The effect of the rotation is also to increase the overall level of soil organic
matter above that which could be produced by only manuring the soil. Partly this
occurs because of the increased yields and better returns of soil organic matter and
partly because of the presence of a forage crop in the rotation. There is a substantial
reduction in erosion (Table 8.4), more than can be explained by averaging out the
erosion on the various crops, and also the length of time that the soil is left bare
and exposed is reduced. Also there is more time over which plant roots are present
to take up nutrients and prevent them being lost; the greater evaporation from the
plants will also reduce the leaching volume. The rotation may not be as good as the
permanent grass cover in preserving soil fertility but provides a larger amount of
food than any other system and reduces soil erosion and nutrient leaching below
anything but a permanent grassland cover.
Table 8.4. Effect of cropping system on erosion and runoff of a silt loam soil
Runoff % Soil loss t/ha Relative loss
grass=1
Fallow 30.7 93.2 122
Maize 29.4 44.1 58
Wheat 23.3 22.6 30
Grass 12.0 0.76 1
Maize/wheat/clover rotation 13.8 6.22 8
Average of maize/wheat/grass 21.6 22.6 30
Source: Miller and Krusekopf 1932
A Lack of Knowledge?
When the experimenters began in the nineteenth century their findings helped to
explain some of the existing practices, but also indicated serious shortfalls in the
availability of materials with which to fertilise the soil. In many cases relatively
simple changes made great improvements to output. This is however not to demean
the accumulated experience of land managers, who in many cases were operating
remarkably efficiently. The extent to which the soil had been altered has only be-
come clear to us in the last century when it was realised that many of the arable
soils probably had less than one-third of the organic matter that they had contained
before woodland clearance. This also demonstrates the extent to which soils can
continue to produce valuable outputs with severely altered properties.
237
Nutrient Flows in Pre-Modern Agriculture in Europe
only be maintained by manures and ash from non-arable land the deterioration
was transferred to these other soils with the arable land struggling along with
lower inherent fertility until the inclusive rotations and use of external fertilising
materials and lime from the eighteenth century ad. The use of terms to describe
land as in-by, which received animal manures, and out-by, which did not, and
the different uses of the soils, indicates the flow of fertility within the farming unit
with the land near the homestead being enriched relative to the outlying areas.
The differences between the in-by and the out-by soils would become considerable
and many years after the nutrients had been applied the phosphorus, which tends
to be best retained by the soil, could still have a beneficial effect on crop growth
which could not be achieved by the use of fresh fertilisers on formerly unfertilised
soils (Roscoe 1960). There is also a trade-off aspect for the land user, with repeated
reuse of land, and therefore soil deterioration, being easier in the short term than
using a larger area less intensively. At a larger scale of distance the growing urban
areas were centres towards which nutrients flowed and where the wastes created
a ring of more fertile soils up to several kilometres wide around the area. Later
urban sprawl has engulfed many of these fertile locations which retain only the
place name Covent and Hatton Gardens in London for example to indicate
the former use. All crop rotations involve both an exploitation phase and a recovery
phase. The ultimate owner of the land must also be considered, for landlords may
constrain the activities of their tenants so as to prevent long-term deterioration of
the soil this is seen in the constraining clauses in many tenancy agreements which
prevented the sale of hay and straw and even prescribed a particular rotation of
crops. The ploughing up of permanent grassland was commonly banned without
payment of a large fine.
The outsourcing of grain by Athens and Rome demonstrates the effect of
decreasing soil productivity in these areas. That grain was then imported from Egypt,
where the soil fertility was maintained by erosion in Ethiopia and redeposition in
the lower Nile valley, and elsewhere, demonstrates the scale of the reduction in
soil productivity in Greece and Italy and the need for their urbanised societies to
have access to products from a large area where the soils had not been so severely
degraded. Although Athens retained a substantial area of fertile soil in the valleys,
where much of it had been redeposited after erosion from the hillsides, this rejuve-
nation was not, as in Egypt, an on-going process so that the fertility of these areas
decreased. To what extent major population movements were caused by the need
to seek new soil can be debated but it is clear that the productivity of considerable
areas of Europe was seriously reduced long before the new Agricultural Revolution
of the seventeenth and eighteenth centuries ad brought more sustainable systems
of management. The benefits to the properties of soils provided by these new rota-
tions the soil organic matter content and structure were improved and, where
necessary, drainage was carried out, lime was applied to raise the pH, urban wastes
239
Nutrient Flows in Pre-Modern Agriculture in Europe
were recycled and bones and basic slag were used to raise the phosphate status
were to enable more people to be fed better off a smaller area of land. Better
and faster plant growth also dried out the soils to greater depth, provided better
cracking of the subsoil and returned larger amounts of roots, increasing soil organic
matter and biological activity in the soil while reducing the risk of soil erosion.
These changes raised the general fertility base, reducing between-soil differences
and giving greater flexibility of use of a wider range of soils. To some extent these
changes were remedying the adverse changes of the previous land use, although the
now-limed sandy podzols would probably have been too acidic for agricultural use
before farming became widespread in north-western Europe. Care must be taken
not to over-interpret the extent of deterioration through exploitation, for some of
the earlier land abandonment had been as a result of climatic change, though even
the nineteenth-century amelioration of the climate did not lead to some of the
degraded areas, for example the North Yorkshire Moors, being reused, even though
this was now technically feasible. In areas of Europe which were less economically
developed traditional practices remained, and the only redeeming feature in these
areas was that the management units were small and tended to integrate livestock
grazed on surrounding hills, so that a supply of nutrients was available in manure.
Much of the more intensively managed land in such areas, for example Dalmatia,
was in basin sites, which received run-on of water and eroded soil from the de-
graded uplands (Chapman et al. 1996). There was therefore a continuation of the
decrease in cultivable area, due to soil loss from the slopes, which had begun before
Plato commented on it (Lee 1971) and it continues today. The more-sustainable
rotational farming system was to be swept aside over large areas by the expansion
of industrial agriculture, particularly after the Second World War. Manufactured
fertilisers, often subsidised, and with food prices buoyed up to maintain farm output
led to maximisation of yield and intensification of use of the land. The larger units
of management that resulted to fit with the enlarged machinery led to soil erosion
(Wicherek 1993) and to widespread concerns over ecology and environment. The
increasing interest, more recently, in organic farming has encouraged a return to
the rotational systems of land use, at a cost to total output, and is probably leading
to an amelioration of soils similar to that seen after the introduction of rotations
in the seventeenth and eighteenth centuries.
References
Anderson, J. 1794. General View of the Agriculture of the County of Aberdeen. Edinburgh:
Blackwood.
Arnon, I. 1972. Crop Production in Dry Regions. London: Leonard Hill.
Ash, H.B. 1941. Lucius Junius Moderatus Columella on Agriculture, vol. 1. London: Hein-
emann
240
Robert S. Shiel
Askew, G.P., R.W. Payton and R.S. Shiel. 1985. Upland Soils and Land Clearance in Britain
During the Second Millennium bc. In D.Spratt and C. Burgess (eds), Upland Settlement
in Britain: The Second Millennium bc and After, BAR British Series 143, 533.
Bear, F.E. 1931. Soil Management, 2nd edn. New York: John Wiley.
Brown, J.R. 1994. The Sanborn Field Experiment. In R.A. Leigh and A.E. Johnston
(eds), Long-Term Experiments in Agricultural and Ecological Sciences (Wallingford: CAB
International), pp. 5382.
Bryson, R.A. and T.J. Murray. 1977. Climates of Hunger: Mankind and the World's Changing
Weather. Madison: University of Wisconsin Press.
Campbell, B.M.S. 1991. Land, Labour, Livestock and Productivity Trends in English Seign
orial Agriculture, 12081450. In B.M.S. Campbell and M. Overton (eds), Land, Labour
and Livestock (Manchester: Manchester University Press), pp. 14482.
C.E.C. 1985. Soil Map of the European Communities 1:1000000. Luxembourg: Commission
of the European Communities.
Chapman, J.C., R.S. Shiel and S. Batovic. 1996. The Changing Face of Dalmatia: Archaeological
and Ecological Studies in a Mediterranean Landscape, Reports of the Research Committee
of the Society of Antiquaries of London No. 54. Leicester: Leicester University Press.
Dimbleby, G.W. 1962. The Development of British Heathlands and Their Soils, Oxford Forestry
Memoirs, 23. Oxford: Clarendon Press.
Eyre, S.R. 1968. Vegetation and Soils 2nd edn. London: Edward Arnold.
Ferro, R.D. 1949. Warping at Blacktoft, 1948. Journal of the Yorkshire Agricultural Society,
5665.
Flint, R.F. 1971. Glacial and Quaternary Geology. London: Wiley.
Glennie, P. 1991. Measuring Crop Yields in Early Modern England. In B.M.S. Campbell
and M. Overton (eds), Land, Labour and Livestock (Manchester: Manchester University
Press), pp. 5177.
Grace, P. and J.M. Oades. 1994. Long Term Field Trials in Australia. In R.A. Leigh and
A.E. Johnston (eds), Long-Term Experiments in Agricultural and Ecological Sciences
(Wallingford: CAB International), pp. 5382.
Gustafson, A.F. 1941. Soils and Soil Management. New York: McGraw Hill.
Hall, A.D. 1905. The Book of the Rothamsted Experiments. London: John Murray.
Halstead, P. and J. Tierney. 1998. Leafy Hay: An Ethnoarchaeological Study in NW Greece.
Environmental Archaeology, 1: 7180.
Jenkinson 1988. Soil Organic Matter and its Dynamics. In A. Wild, Russells Soil Conditions
and Plant Growth (Harlow: Longman), pp.564607.
Lee, H.D.P. 1971. Platos Timaeus and Critias. Harmondsworth: Penguin Books
Lefevre, T., J.L. Todoc and G.R. Timilsina. 1997. The Role of Wood Energy in Asia. http:
//www.fao.org/docrep/w7519e/w7519e00.htm
Leigh, R.A. and A.E. Johnston (eds). 1994. Long-Term Experiments in Agricultural and
Ecological Sciences. Wallingford: CAB International.
Leonard, W.H. and J.H. Martin. 1963. Cereal Crops. London: MacMillan.
241
Nutrient Flows in Pre-Modern Agriculture in Europe
Lockwood, J.G. and J.G. Sellers. 1983 Some Simulation Models Results of the Effect of
Vegetation Change on the Near-Surface Hydroclimate. In A. Street-Perrott, M. Beran
and R. Ratcliffe (eds), Variations in the Global Water Budget (Dordrecht: Reidel), pp.
46377.
McConkey, B.G., C.A. Campbell, R.P. Zentner, F.B. Dyck and F. Selles. 1996. Long-Term
Tillage Effects on Spring Wheat Production on Three Soil Textures in the Brown Soil
Zone. Canadian Journal of Plant Science, 76: 74756.
Miller, M.F. and H.H. Krusekopf. 1932. The Influence of Systems of Cropping and Methods of
Culture on Surface Runoff and Soil Erosion, Research Bulletin 177. Missouri Agricultural
Experimental Station.
Moore, H.I. 1944. Grassland Husbandry. London: George Allen & Unwin.
Parry, M.L. 1978. Climatic Change, Agriculture and Settlement, Studies in historical geography.
Connecticut: Archon Books.
Rackham, O. 1982. Land Use and the Native Vegetation of Greece. In M. Bell and S. Lim-
brey (eds), Archaeological Aspects of Woodland Ecology, BAR (Int Ser) 146: 17798.
Renfrew, C. 1987. Archaeology and Language: The Puzzle of Indo-European Origins. London:
Jonathan Cape.
Renfrew, J.M. 1973. Palaeoethnobotany. London: Methuen.
Roscoe, B. 1960. The Distribution and Condition of Soil Phosphate under Old Permanent
Pasture. Plant and Soil, 12: 1729.
Russell, E.J. 1915. The Fertility of the Soil. Cambridge: Cambridge University Press.
Shiel, R.S. 1991. Improving Soil Fertility in the Pre-Fertiliser Era. In B.M.S. Campbell
and M. Overton (eds), Land, Labour and Livestock (Manchester: Manchester University
Press), pp. 5177.
Shiel, R.S. 1997. Surviving in the Tisza Valley; Plants and Peoples Perspectives on the Envi-
ronment. In J.C. Chapman and P. Dolukhanov (eds), Colloquia Pontica 3, Landscapes in
Flux: Central and Eastern Europe in Antiquity (Oxford: Oxbow books), pp. 18192.
Shiel, R.S. 2002. The Future of Farm Grassland as a Diverse Productive Environment.
Transactions of the Natural History Society of Northumbria, 62: 11522.
Shiel, R.S. and J.C. Batten. 1988. Redistribution of Nitrogen and Phosphorus on Palace
Leas Meadow Hay Plots as a Result of Aftermath Grazing. Grass and Forage Science,
43: 10510.
Shiel, R.S. and Chapman, J.C. 1988. The Extent of Change in the Agricultural Landscape
of Dalmatia, Yugoslavia, as a Result of 8,000 Years of Land Management. In J.C.
Chapman, J. Bintliff, V. Gaffney and B. Slapsak (eds), Recent Developments in Yugoslav
Archaeology, BAR Int Ser 431: 3144.
Shiel, R.S. and Stewart, A. 2003. The Soils and Agricultural Potential of the Thespiai
Area. Unpublished report of the CambridgeDurham Boeotia Project, available from
R.S.Shiel@ncl.ac.uk
Sillitoe, P. and Shiel, R.S. 1999. Soil Fertility under Shifting and Semi-continuous Culti-
vation in the Southern Highlands of Papua New Guinea. Soil Use and Management,
15: 4955.
242
Robert S. Shiel
Smith, L.P. 1976. The Agricultural Climate of England and Wales, MAFF Technical Bulletin
35. London: HMSO.
USDA, 2003, Food and Nutrition Information Center. http://www.nal.usda.gov/fnic/
van Andel, T.H. 1987. The Landscape. In T.H. van Andel, and S.B. Sutton (eds), Landscape
and People of the Franchthi Region (Bloomington: Indiana University Press), pp. 364.
West, M.L. 1988, Theogony and Works and Days. Oxford: Oxford University Press.
Whitehead, D.C. 2000. Nutrient Elements in Grassland: SoilPlantAnimal Relationships.
New York: CABI.
Wicherek, S. (ed.). 1993. Farm Land Erosion in Temperate Plains Environment and Hills.
Amsterdam: Elsevier.
Wilkinson L.P. 1978. The Georgics of Virgil: A Critical Survey. London: Cambridge Uni-
versity Press.
243
Human Interaction with Soil-Sediment Systems in Australia
R.J. Wasson
Introduction
Most environmental consequences of economic development are unanticipated,
ignored because they are not believed to be significant, or responded to either
by attempts at repair or by adaptation well after they have become plain to see.
Many environmental changes occur well after their human or non-human causes,
making the link between cause and effect difficult to identify. There is therefore a
delay between cause and effect, response, and either improvement or adaptation.
History is essential to understand such a system of interconnected processes and
phenomena (Newell and Wasson 2001) so that changes and delays resulting from
present and future economic development can be anticipated; the causes of current
changes identified; and more timely responses developed.
Environmental history therefore can be of great practical value for present
day natural resource and environmental management (Wasson and Clark 1985,
Wasson 1994, Dovers 2000). River catchment (or watershed) response to land
cover and land use change, and the ways in which people have perceived and re-
sponded to the changes in rivers in particular, are the subject of this chapter. The
erosion and transport of soil and sediments through river catchments provides a
useful case for analysis because, in Australia, European settlement in catchments
produced a shock (or perturbation) that can be identified and is still working its way
downstream. Much of the European history of Australia is recorded, and in some
places former farmers are still able to remember river changes set off by removal
of native vegetation. These people, and others, can also tell us how and why they
responded to environmental change. All of the difficulties of oral history cannot
244
R.J. Wasson
be overcome, but the method of Starr (1989) has been used at different study sites
to at least provide consistency of approach.
rate at the mouth of a river, for example, may reflect a perturbation in the head-
waters decades or centuries before. This idea is extremely important for analysis of
human response to change where delays are large, and the cause of change distant
in space.
A Conceptual Framework
A conceptual framework is needed for analysis of the biophysical changes triggered
by land use and climate change, for the human response to such changes, and for
the biophysical changes caused by human responses.
A Spatial Framework
The first part of this framework is a conceptualisation of the sources, transport and
sinks of sediment in a river catchment (Figure 9.1). The catchment is subdivided
into sub-catchments according to the Strahler method, as follows. Each division is
given an order number, beginning with zero for the sub-catchment upstream of a
channel head, and then subsequently numbered downstream. Channels are divided
into links, so a first-order link is that which begins at the downstream margin of
a zero-order sub-catchment and ends at the first stream junction downstream. A
second-order link is produced by the junction of two first-order links. A third-order
link is produced where two second-order links join. A first-order sub-catchment
is the total area from which water drains to a first-order link. A second-order sub-
catchment is the total area from which water drains to a second-order link; and so
on for higher orders.
Soil is eroded from hill slopes by sheet erosion, rill erosion, gullying, and
landslides. Some of this material reaches a river channel link, and the rest remains on
colluvial footslopes perhaps to be moved during subsequent runoff events. Therefore,
some eroded material goes into a sink (Figure 9.1), at least temporarily. Over long
periods, all such material reaches the channel network. Within the channels, some
sediment is temporarily stored (the in-channel deposits in Figure 9.1), to be moved
later. Secondorder links commonly are flanked by floodplains, where sediment
is stored and rudimentary soil formation occurs. Floodplains are also sources of
sediment as channels migrate laterally, very large floods enlarge channels, or land
use reduces the resistance of stream banks to erosion. Rivers also incise, that is
they cut downward into their own deposits to produce a channel that contains all
flows. This process releases very large amounts of sediment to downstream reaches.
Further downstream, for links greater than third-order, floodplains are wide enough
to prevent soil eroded from adjacent hill slopes reaching the channel. Still further
downstream, sediment reaches a reservoir, lake, estuary or ocean.
Ideally each of the inputs and outputs to the components in Figure 9.1 is
quantified, and a history of change provided in the form of a time-series of fluxes.
246
R.J. Wasson
SOURCES SINKS
hillslope soils
colluvial footslopes
in-channel deposits
hillslope soils 2nd order channel
floodplains
estuary, reservoir,
lake, ocean
Biophysical Change
The second part of the conceptual framework requires a way of including per-
turbations and biophysical responses. Each of the components of Figure 9.1 is
susceptible to change by external forces, such as climate change, large magnitude
rainfall events, and/or land use change. They also change by internal processes such
as by passing a gradient threshold so that an external perturbation that previously
had no effect now has an effect.
One way of envisaging these changes is by means of the diagrams of Figure
9.2. Perturbations can be thought of as ramps, steps and spikes. The ramp (A) may
be the gradual change of land cover as hunting and gathering gives way to patches of
settled agriculture and then large areas of agriculture. The response to this ramped
change may be instantaneous (solid line) in sensitive systems where soils are highly
erodible, or delayed in low sensitivity systems where soils are resistant to erosion.
The delay is equivalent to a threshold. Where soil organic matter and tree roots
persist after clearing, erodibility is low and a threshold (t) caused by organic matter
decay must be passed before erodibility is increased. Another threshold might be
the connection of patches of cleared land sufficient to allow eroded soil to reach a
stream channel, so that the dashed line represents stream sediment load.
The step change perturbation (Figure 9.2B) mostly occurs in places where
industrial agriculture has been introduced very quickly, such as in Australia and
North America. Here the tools were available for rapid clearing of forest, for exam-
ple, producing a near-instantaneous increase in runoff and soil erosion in the case
of sensitive soil. Where organic matter helps the soil to resist erosion, erosion may
increase slowly soon after clearing (Ba), or a threshold (t) of soil organic matter
decay has to be passed before erosion increases markedly (Bb).
Both ramps and steps are representations of changes that do not recover
quickly, such as land cover change. The responses, on the right-hand side of Figure
9.2, are represented as smooth (average) conditions in A and B. In reality, there
will be infrequent large magnitude events superimposed upon the smooth curves.
Such spikes are shown separately in Figure 9.2 C and D for ease of representation.
A single spike (C), such as a large rainfall event, is responded to quickly by soils
on steep slopes that fail by for example rotational slump (Ca). On gentler slopes,
soils may fail by earthflow, and so the response to a single spike of rainfall will be
slower and last longer as a period of movement (Cb).
Multiple spikes produce rapid recovery in highly sensitive systems such as
small catchments where there are large numbers of channels that drain quickly
(Da). A spike of runoff and sediment transport in larger catchments will wane more
slowly because of the larger travel time. If the interval between spikes is shorter than
the recovery time then sediment transport will continue to increase, as depicted in
Figure 9.2(Db), making it appear that there is an increasing cause.
248
R.J. Wasson
A
RAMP
t
time time
B
STEP
a
b
t'
time time
C
SPIKE
a b
time time
D
SPIKES
b
b
a a a
time time
Cause incorrectly
No impact Impact identified along with
a solution
Action
No action
Negative Positive
outcome outcome
Other examples contained in Figure 9.3 include the case of human impact
where the cause of the biophysical change is known but there is no solution that is
economic, technically feasible, politically tractable or socially acceptable. Therefore,
there is no action.
All of the cases described so far from Figure 9.3 explain why no action
occurs for very long periods of time following biophysical change. Particularly in
the last few decades, impacts on non-human organisms have attracted attention.
So along with impacts on people in Figure 9.3, impacts on other organisms could
also be considered. Impacts in Figure 9.3 are therefore diverse.
250
R.J. Wasson
The remainder of Figure 9.3 consists of cases that result in action, for better
or worse. The cause of a biophysical change may be incorrectly identified and a
solution is found that is believed to be viable, leading to either no amelioration of
the biophysical change or a worsened condition. If the cause is correctly identified,
and a solution found that is tractable, then action will lead to a positive outcome.
There may still be a delay between observing a biophysical change and taking effective
action because the cause may be discovered well before a viable solution is found;
or other matters have a higher priority for expenditure and effort; or institutional
arrangements do not permit action.
To Figure 9.3 can be added a fourth component, namely feedback from
action to the biophysical system. If the cause of the initial biophysical change is cor-
rectly identified, and an effective solution adopted, then the damaging biophysical
change may be slowed, stopped, or even reversed. Depending upon the effectiveness
of actions, there may or may not be further investigation and action. If action is
adaptation, then the only continuing biophysical change will be according to the
response trajectory that is determined by the processes and feedbacks within the
biophysical system. In some cases, a biophysical system once perturbed will relax
to a new state or take a new form which may be less problematic for people (and
other organisms). An example is the triggering of gullying by land cover reduction
(Wasson and Sidorchuk 2000). A gully network grows until headcutting stops as
the catchment area to each gully head is reduced to a critical size. Sediment yield
in such a system therefore rapidly increases as the network grows, and then falls
as the network stops growing. Downstream sediment transport also slows, not as
a result of human intervention but as a result of a negative feedback that stabilises
the system.
In summary, the conceptual framework that is adopted in this chapter has
the following components:
a spatial framework (Figure 9.1) based on the hierarchy of the stream net-
work and its sub-catchments, taking explicit account of the sources, sinks
and yields of, in this case, sediment
a temporal framework (Figure 9.2) in which the magnitude of processes
inducing change are represented by a few simple conceptualisations, and the
responses, measured as changes in form (of rivers particularly) and transport
rate of sediment, are conceived as averaged trajectories. If individual events
are known, the curves can be represented more accurately.
the causal links between observations, impact, and action (or no action) in
which delays are implicit (Figure 9.3). Impacts on both people and other
organisms can be treated within this component.
251
Human Interaction with Soil-Sediment Systems in Australia
Figure 9.5. Location of the soil conservation projects in the Molonglo Catchment
rabomberra Creek much as they are today; that is, vertical walled and incised. The
diary account is by John William Buckle Bunn, a remarkable teenager living on
Woden Station. His sketches and descriptions show that incision of Woden Creek,
a tributary of Jerrabomberra Creek, had occurred a little before his diary entry. This
conclusion is evidenced by his sketch of tree roots protruding near the top of the
254
R.J. Wasson
gully well. A well had been sunk in the bed of the gully, the remains of which can be
seen today (B Starr, pers. comm.). There is no evidence however of any response to
this incision of Woden Creek. It is not until the 1950s that attempts were made to
control stream-bank erosion by planting poplars and willows, on the middle reaches
of Jerrabomberra Creek. At one particularly active bend, car bodies were piled at
the base of a channel wall and poplars planted. The car bodies are now stranded in
mid-stream, showing that they had no effect on stream-bank erosion.
For most of the history of farming in the Molonglo catchment, farmers
concentrated their efforts on productive land. They introduced non-native grasses,
a process known as improving pasture, managed stocking rates, fenced land,
established dams, controlled feral animals, reduced weeds as best they could, and
benefited from the reduction of rabbit numbers by the spread of the myxoma virus
in the 1950s (Starr et al. 1999).
It is likely that farmers noticed the incision of valley floors and the destruc-
tion of valley floor swamps, because they went to some trouble in some places to
build bridges where once crossings of valley floors were simple matters (Starr, et al.
1999). But apart from planting a few trees they either did not have the resources
or the motivation to do more to stabilise the channels. It seems that the farming
community adapted to the changed valley floors.
cases. The NSW Water Conservation and Irrigation Commission had authority
over streams, but attempts to involve them in the Molonglo catchment broke down
in 1973 (Starr 1991).
While the significance of stream-bank erosion had been recognised as early
as 1958, its control had thus far been indirect; and it is not at all clear that peak
flows in large erosive events were reduced by structural works because there are no
flow data for most of the catchment. The views of local landholders on these matters
are not recorded, except for a report in the Canberra Times (10 May 1966) of the
concerns of Mr Bernard Morrison, a local grazier, which have been interpreted to
refer to stream-bank erosion not being under control (NSW Department of Land
and Water Conservation 2000).
The Protection Scheme continued in stages (Figure 9.5) with most emphasis
on sheet erosion and gully stabilisation until quantitative sediment sourcing studies
began to alter the thinking of the NSW Soil Conservation Service. By estimating
the erosion rate by sheet processes from different land cover types, from gullies,
Lessons had also been learned about broad-acre soil conservation (Fogarty
et al. 1989). Absorption banks reduced access across paddocks. Grass flumes often
failed because grass cover was difficult to maintain in a climate with a short grow-
ing season. Dams on larger gullied catchments could now be built by using PVC
pipes as primary spillways, because earth spillways, where flows are large, often fail.
Finally, dams were being located more strategically to catch water and sediment from
ploughed paddocks, roads and gullies. It is clear that the NSW Soil Conservation
Service had been pushed very hard by the technical difficulties of the Protection
Scheme, and innovation had occurred throughout the Scheme. Over 154 km of
banks had been built and over 226 km of gullies shaped or filled. Nearly 1000
gully control structures or dams had been built, often at the expense of landhold-
ers. Not only had erosion been slowed but land was now more profitable (NSW
Department Land and Water Conservation 2000).
The Commonwealths interest in the Scheme came to an end in 1998 as
it was clear that most continuing effort would mostly benefit landholders, and all
that could be done to save Lake Burley Griffin from sedimentation had been done.
Also, the Scheme was out of step with total catchment management objectives of
solving multiple problems rather than just erosion and sedimentation. Revegetation
has become a priority, often planting trees where they did not exist naturally!
The environmental history of soil erosion and conservation in the Molonglo
Catchment is summarised in Table 9.1 within the conceptual framework con-
structed earlier. The spatial framework used by farmers appears to have been their
own properties, and soil conservators thought of the catchment as a collection of
properties rather than as a hierarchical nesting of catchments as in Figure 9.1. The
hierarchical view, with connectivity between hill slopes and channels, and between
channels, only became clear in the 1980s and 1990s as scientists interacted with
soil conservators. Treatment under the Protection Scheme of important sediment
sources only became focused when a sediment budget was available.
There is little detail of Perturbation and Biophysical Responses, but it can be
inferred that grazing of valley floors led quickly to incision and export of sediment.
While there is no documentation, it is likely that observant farmers would have
seen sheet erosion once stocking rates became high in the late nineteenth century.
It is not clear that they did anything about this situation until rabbit numbers were
reduced by the myxoma virus and improved pasture became possible in the 1960s
with the aid of fertilisers.
People responded to incision by adaptation, although there were a few
attempts at channel stabilisation before the Protection Scheme. But most farmers
focused on their properties, and it was not until a top-down government program
of catchment protection was developed that landholders became interested in larger
areas. While this conclusion is reached largely by inference, the few remaining farm-
ers of sufficient age support this conclusion. Soil conservators altered their views of
259
Human Interaction with Soil-Sediment Systems in Australia
Table 9.1. Summary of soil erosion and conservation in the Molonglo catchment
Spatial framework Perturbations and Causation and responses,
biophysical responses and feedback
Priority given to sub-catch- Grazing of valley floors led Early recognition of
ments badly eroded by sheet to rapid incision (as a step incision but no effective
processes. Other processes change, Figure 2). Delay be- control until 1990s; a delay
noted. No explicit attention tween grazing and incision of at least 30 years. Most
paid to connectivity. Late in probably less than 10 years. responses were to sheet
the Scheme, quantification Sheet erosion probably and gully erosion for both
of sediment sources led to became obvious soon after technical and institutional
more explicit treatment of heavy stocking in the late reasons. Soil conservators
particular locations of high nineteenth century. adapted their methods both
sediment yield. as they learned from their
traditional techniques, and
as they focussed on stream-
bank erosion.
de-vegetated and Forrests fine grassy plains were bare. The prevailing leasehold
policy of subdivisions based on river frontages actively discouraged development
of permanent watering points away from the rivers (Bolton 1953). Yet cattle had
managed to de-vegetate large areas away from the rivers, as well as along them;
albeit in a time of drought.
Teakle (1944) and Medcalfe (1945), two agricultural scientists, reported
massive erosion. They saw exposed tree roots, soil pedestals, and an absence of
A-horizon (topsoil) over large areas. Sheet and rill erosion of between 15 and 30
cm was estimated. This amounts to 3,500 to 7,000 t/km2/year, a very high rate
of erosion.
Forrest noted gullies along the Behn River frontage because he nearly fell
into them, but he did not record other gullies either along or away from rivers.
Today gullies are common, particularly across the Hardman Syncline and also along
the frontages of most rivers. Teakle and Medcalfe focused on sheet and rill erosion
presumably because of their interest in the pastoral enterprise. Medcalfe, however,
concluded that the serious erosion he observed in the Hardman Syncline must be
producing excessive sedimentation downstream.
In 1917, the Ord River Station, a cattle property first occupied by N. Bucha-
nan in 1884, was subdivided into seven stations (Pratchett 1990). Impetus was
given to the cattle industry in 1918 by the opening of a meat works at Wyndham
and the beginning of live cattle export. Grazing pressure probably increased as a
result of both subdivision and increased access to markets.
The irrigation potential of the alluvial soils in the vicinity of the modern
town of Kununurra was early recognised. Agricultural experiments between 1941
and 1945 by a member of the pioneering Durack family (Durack 1945) were
promising, and initial exploration for dam sites by the Director of Works Western
Australia occurred in 1941 (Dumas 1944).
Durack also recognised that uncontrolled grazing by cattle leads to pasture
destruction and erosion. But it was not until the 1950s that a systematic survey of
erosion occurred (Stewart et al. 1970). Meanwhile, the idea of further development
of north-western Australia, either for pastoral use or irrigated agriculture, would
serve no useful purpose (Ord River Irrigation Area Review 1978). But factors
other than economics were at play. Decisions taken after the Second World War
were made when the threat of Asian invasion had just been shown to be possible.
The defeat of the Japanese Imperial Army in New Guinea had been a close-shave,
and, for many, demonstrated the correctness of the earlier policy of populating the
north because if we did not the Asians would. In addition, the developmental and
political ambitions of the Western Australian government drove plans for irrigation
on the Ord River.
As the bureaucratic and political arguments about a dam and the development
of irrigation continued, there is no evidence that stocking rates declined. Fencing
261
Human Interaction with Soil-Sediment Systems in Australia
occurred, allowing the control of stock locations, and bores were sunk to provide
water away from the river frontages. This may have reduced stocking pressure near
the rivers, but the damage in the form of gullying had already occurred before the
1960s when the first high quality photographs of erosion phenonoma were taken
by officers of the WA Department of Agriculture (A. Payne, pers.comm.). Written
descriptions from a decade before (Stewart et al. 1970) suggest that gullying had
not markedly changed from that photographed in the 1960s.
In 1959 the decision was taken to build a Diversion Dam near Kununurra,
prior to construction of the main Ord River Dam (Patterson 1965). The alarming
262
R.J. Wasson
reports in 1944 of soil erosion upstream led to revegetation of part of the visibly
worst degraded area in the Hardman Syncline.
This response was thought to be essential to save the Diversion Dam from
sedimentation. Large areas of the soil of the Hardman Syncline (Figure 9.8) had
been de-vegetated by cattle. The soil surface rapidly sealed as pores infilled with
particles washed across the surface and rain splash puddled the surface. Infiltration
by rainfall into the soil was reduced and revegetation by natural means was almost
impossible. This sequence of events is common throughout the worlds semi-arid
tropics and sub-tropics (Walker et al. 1981). There is generally a ten-fold reduc-
tion of infiltration rate when a grass-litter cover is removed to create bare soil. This
change is supported by Fitzgeralds (1968) infiltration measurements in the area
that became known as the Ord River Regeneration Reserve (ORRR) where most
of the revegetation occurred.
Mechanical pitting of the soil surface by means of tools pulled by tractors
was necessary to break the surface seal, along with direct seeding into the pits. Exotic
species were used because they produced more seed than the native species. Absorp-
tion banks were also built to slow runoff and thereby reduce sheet erosion, and to
trap seeds and speed regeneration in areas away from lines of pitting. Re-seeding
was restricted to areas that could be reached by tractors, limiting it to relatively flat
areas between gullies. Some attempts were made to re-seed gully walls, where they
were gentle, and some seed reached gully floors resulting in limited revegetation.
But there is no evidence of an explicit attack on gully erosion. Once again, the
agricultural scientists involved in this superhuman revegetation program believed
that reducing runoff from areas between gullies by revegetation would slow gully
erosion. (A. Payne, pers. comm.).
In 1964 the Western Australian government continued to lobby the Com-
monwealth government for financial assistance with the Ord Irrigation Scheme. By
1965 there was a new argument for building the Main Dam; to slow sedimentation
of the Diversion Dam (Patterson 1965). Revegetation of the most severely eroded
land was assisted in 1968 by resumption of the land by the Western Australian De-
partment of Agriculture, and the creation of the ORRR. While the active reseeding
program was largely finished by this date, the resumption allowed the Department
to control stocking rates and thereby maintain vegetation cover.
Funding for the Main Dam was approved and the reservoir began to fill in
1972. Measurement of sediment concentrations and discharges in the Ord River by
the Western Australian Public Works Department were used to estimate the annual
average sediment transport rate to the reservoir at 29.768.60 x 106 t/yr (Wasson
et al. 2002). The large uncertainty of 8.60 x 106t/yr, which is 29 per cent of the
average, is entirely the result of interannual variability of flow rates and therefore
sediment transport rate. Errors due to sampling are unknown.
263
Human Interaction with Soil-Sediment Systems in Australia
In 1985 a survey of the total amount of sediment in Lake Argyle was carried
out by the Western Australian Water Authority (Wark 1987), using coring, re-survey
of cross sections established when the dam was built, and echo-sounding. On this
occasion measurement errors were estimated, giving a result of 23.54.7x106 t/yr
for the average sedimentation rate. The two figures are statistically identical, and
R.J. Wark, an insightful engineer, calculated that revegetation of the Hardman
Syncline had not been successful.
This conclusion led to a sediment tracing study that attempted to determine
if revegetation of the soft rocks and erodible soils of the Hardman Syncline was the
correct strategy. Mineral particle magnetic properties, uranium series nuclides, and
rare earth isotopes, were used to determine which sub-catchments produced most
sediment. In addition, the proportion of topsoil in river sediments was determined
measuring the amount of 137Cs in Lake Argyle sediments, a radioactive nuclide that
occurs in topsoils across the planet. Two major conclusions were reached (Wasson
et al. 2002). First, the Hardman Syncline, and similar geomorphic zones nearby,
produced most of the sediment in Lake Argyle. Second, most of the sediment was
produced by gullying and channel erosion, not by sheet and rill processes. These
conclusions meant that while the revegetation project had targeted the right part
of the catchment, the major erosion process had not been substantially affected.
While the tracer study was going on, officers of the Western Australian
Department of Agriculture had independently recognised that gullies were an im-
portant source of sediment. They had built small brushwood and bamboo fences in
some gullies, but they were washed away during the next wet season. Revegetation
of the gullies by an expansion of the 1960s scheme was not seriously considered,
and so the ORRR was destocked of both domestic and feral cattle. The low value
of the land, and a shift to a providerprovidee method of internal allocation of
funds (i.e. the need for any research project to have an internal user other than the
proponent of the research) within the Western Australian bureaucracy made any
other course of action impossible. However the Department plans to re-survey the
sediment in Lake Argyle and examine the gullies in the ORRR to determine if any
change has occurred.
In summary (Table 9.2), the problem of sedimentation only came to promi-
nence once plans were underway for a large irrigation reservoir. Soil erosion had
been commented upon earlier by agricultural officers but there is little evidence
that pastoralists did much to ameliorate it. Fencing and provision of waters away
from rivers probably reduced grazing pressure and therefore erosion along river
frontages, but these measures were almost certainly designed to improve stock
management rather than control soil erosion.
With a reservoir being planned, scientists focused attention on the area most
badly eroded. This approach is essentially the same as that used in the catchment of
Lake Burley Griffin. There was no attempt to quantity sediment sources, rather they
264
R.J. Wasson
Table 9.2. Summary of soil erosion and conservation in the Ord River catchment
Spatial framework Perturbations and Causation and responses,
biophysical responses and feedback
Priority for conservation Heavy grazing by cattle Early recognition of
given to the Hardman Syn- of grasses with low seed erosion, but no effort at
cline, but emphasis given to setting rate in a semi-arid control until a reservoir was
sheet erosion. Quantifica- monsoonal environment envisaged. Large scale reveg-
tion of sediment sources produced massive erosion. etation of the most eroded
confirmed for location but Decreased infiltration pro- area. Later assessment
not the assumed dominant duced a de-vegetated condi- showed that the scheme was
erosion process. tion. Increased downstream not focussed on the main
sediment transport erosion process, gullying.
No further attempts at ac-
tive erosion control; rather,
the most eroded area has
been de-stocked.
relied simply upon field survey and description. An enormous effort at revegetation
was based upon a very limited understanding of erosional processes and rates. The
role of local pastoralists in this venture appears to have been slight, all of the effort
and cost being borne by the Western Australian government.
By the 1980s R.J. Wark decided that the effectiveness of the revegetation of
the ORRR needed to be determined. This is a rare, and early, case of a government
agency testing the effectiveness of an intervention. The conclusion from re-survey
of the amount of sediment in Lake Argyle led to an attempt to quantify sediment
sources because it was possible that revegetation had occurred in the wrong place.
The tracer study showed that revegetation was spatially appropriate but had not paid
sufficient attention to the main erosional process, gullying. Once this was known,
the response was swift and appropriate to the institutional setting of the time. The
ORRR was destocked and left to revegetate by natural means.
west and south, grassland south of the Stirling Ranges, and mallee (multi-stemmed
eucalypts) woodland and tall scrubland in the eastern part of the catchment. About
72 per cent of the catchment has been cleared, mainly for sheep and cattle grazing,
cereal cropping, and limited viticulture (Seal 1995).
European exploration of the area began seriously in the 1820s and 1830s.
Most early settlement was along the coast, but by 1840 settlers had spread from
Albany along the Kalgan River (Blight 1996). Major land clearing occurred after the
Second World War, largely at the instigation of the government to settle returning
soldiers and to open up the country for economic development (Gaynor 2002).
266
R.J. Wasson
Phosphatic fertilisers were an important part of the agricultural system, given the
low natural nutrient status of the soils of the catchment.
In 1961 the seagrass of Oyster Harbour was surveyed and found to be in
good condition (Seal 1995). Subsequently, macro algae began to appear in the estu-
ary, and seagrass was smothered. When resurveyed in 1981, the seagrass biomass
was only 23per cent of that in 1961. By 1992 it was only 10per cent of that in
1996, and macro algal biomass had increased 14-fold. This fundamental ecosystem
shift was attributed to excessive quantities of phosphorus reaching the estuary from
eroded paddocks where fertilisers had been applied.
The Environmental Protection Authority (of Western Australia) monitored
nutrient loads starting in 1988, but it was not until the 1990s that sufficiently
detailed measurements showed that about 70per cent of the total phosphorus
was particulate, and the rest dissolved (Weaver, et al. 1994). This may even be an
underestimate of the particulate fraction. Therefore, erosion of phosphorus-enriched
soil particles is clearly the most likely source of the nutrient that was damaging the
Oyster Harbour ecosystem. Sedimentation of the harbour was also noted by local
fishermen (the West Australian newspaper, 4 January 1994).
In 1990 the Environmental Protection Authority recommended that nutri-
ent input to Oyster Harbour should not exceed its assimilation capacity. Targets for
each source of phosphorus were established, and management plans constructed.
Importantly, scientific research by the Western Australian Department of Agri-
culture began at the same time as, rather than decades after, a problem had been
recognised. In this way, the Kalgan differs markedly from the Lake Burley Griffin
and Ord cases.
The management response consisted of determining the phosphorus needs
of crops and pastures and tailoring fertilisation to these needs, and planning the re-
establishment of riparian vegetation both to intercept runoff of soil from paddocks
and to slow stream bank erosion. These measures were targeted particularly to low
order channels because they are by far the largest proportion of channels, and soil
can reach them most easily. This is the only example in our three case studies where
the stream order hierarchy has been used in management interventions.
The management planning has been aided by a total phosphorus budget for
the catchment (Weaver et al. 1999). From the total channel length in the catch-
ment, an estimated average riverbank erosion rate and bank height, and an assumed
100per cent delivery ratio (that is, all mobilised phosphorus reaches Oyster Har-
bour), Weaver estimated that 23per cent of the annual average phosphorus load
reaching the estuary could be from riverbank erosion. Rivers in this case include all
channels including gullies. This calculation probably overestimates the amount of
phosphorus coming from stream banks because the delivery ratio is almost certainly
less than 100per cent and the assumed bank erosion rate too high. Nonetheless,
to know that more than 77per cent of particulate phosphorus in the estuary is
267
Human Interaction with Soil-Sediment Systems in Australia
coming from surface erosion, mostly off agricultural land, has been of enormous
value to the catchment managers and farmers in the Kalgan area.
So far emphasis has been given mainly to the observations and responses of
government agencies and officers. What of the farmers? Systematic interviewing of
farmers in small sub catchments in the vicinity of Takalarup supply the only avail-
able information (Figure 9.9). The four sub catchments are different from most
of the rest of the Kalgan catchment because here channel incision has been more
pronounced since significant clearing along valley floors began in about 1907.
Incision of Takalarup Creek began in 1939 as the result of an intense rainstorm on
cleared land. Gullies were also formed in the area by the same event. None of the
present or former landholders spoke of attempts to control incision or gullying.
Rather they adapted by building bridges and by fencing. Salt scalds existed in 1949
in parts of these catchments, and attempts were made to control their spread and
to slow the attendant erosion by using straw. This was of limited success.
Sediment budgets for Takalarup Creek show that between 1907 and 1955,
80per cent of sediment yield came from channel incision. As clearing spread up
slopes away from valley floors, the sheet erosion component of the budget increased
so that channel erosion produced only 38per cent of sediment yield between 1955
and 1997.
The farmers in this area, like those throughout the Kalgan catchment, are
engaged in riparian vegetation reconstruction, minimum tillage and careful use
of fallowing. While there is some evidence that the changed cultivation practices
reflect a desire to maintain productivity in the face of both water and wind erosion,
riparian zone reconstruction is a response to eutrophication of Oyster Harbour.
Large-scale revegetation to reverse dryland salinity remains a largely uneconomic
proposition. Interviews with farmers show that they are now living with saline scalds
and seeps, and salinised streams. Some are growing saltbush (Atriplex spp) on areas
of saline discharge to both reduce discharge and to feed sheep. These responses are
the product of pragmatism and the nationwide Landcare movement, a government-
assisted program involving local people in natural resource management.
It is too early to judge the effectiveness of all of these responses to land
degradation. Reversal of eutrophication should be possible, although anoxic con-
ditions in bottom waters of Oyster Harbour will continue to release phosphorus
from sediments on the bed for decades to come. Reversing sedimentation is a much
more difficult proposition.
Adaptation and local but minor attempts at amelioration of land degradation
occurred before a catchment-wide solution was necessary to reverse eutrophica-
tion of Oyster Harbour (Table 9.3). Advice from Western Australian government
agencies before the introduction of catchment management was largely focused
on farm productivity, and certainly progress with controlling wind erosion in
particular has been made.
268
R.J. Wasson
Table 9.3. Summary of soil erosion and eutrophication in the Kalgan River
catchment
Spatial framework Perturbation and Causation and responses,
biophysical response and feedback
Until eutrophication was Land clearing and ferti- Local erosion was adapted
an issue, farmers focussed lisation caused increased to, while local salinisation re-
on their farms. Catch- sedimentation and eutrophi- ceived some local attention.
ment-wide thinking cation of Oyster Harbour, Once eutrophication was
introduced the idea of and infilling of pools in recognised, farmers worked
stream order for manage- major rivers together to reduce the trans-
ment priorities port of soil and phosphorus
from their farms, with both
economic (on farm) and
environmental benefit.
Conclusions
In all three case studies, changes to the erosionsedimentation system were ob-
served by local farmers and inhabitants to varying degrees. Almost all responses
were adaptive, involving bridges and fencing to keep stock safe and allow access
on farms now divided by deep gullies. A few instances of stream-bank protection
occurred in the Molonglo catchment, but in the Ord the size of the individual
farms and the cost of such measures militated against their use. In the Kalgan
catchment, stream banks of high order channels were protected by not clearing
riparian vegetation, but this was probably because the agricultural benefit of clear-
ing would have been minimal.
It was only when a problem was identified that required a catchment-wide
management response that farmers adopted a wide spatial perspective. In the
269
Human Interaction with Soil-Sediment Systems in Australia
the Molonglo and Ord catchments were either before or during the early years
of catchment management in Australia. Management of the Kalgan has therefore
benefited from this development.
The conceptual framework set up at the beginning of this chapter has
been used where possible to analyse the three case studies. But most of the data
necessary to use the framework properly and fully do not exist. The framework
nonetheless stands as a device for others with more data, and possibly for structur-
ing new studies.
Acknowledgements
For collaboration, many fieldtrips, and stimulating discussions, I wish to thank Barry Starr,
Jon Olley, David Weaver, Paul Novelly, and Graham Clifton.
References
Blight, J. 1966. South Coast Regional Initiative: Land and Water Care Strategy. The Great
Southern Development Commission and Agriculture Western Australia
Bolton, G. 1953. A Survey of the Kimberley Pastoral Industry from 1885 to Present. MSc
Thesis, University of Western Australia.
Dovers, S. (ed.) 2000. Australian Environmental History and Policy: Still Settling Australia.
Oxford: Oxford University Press.
Dumas, R.J. 1944. Development of the North-West of Western Australia. Journal of the
Institution of Engineers of Australia, 16: 6570.
Durack, K.M. 1945. The Monsoonal North of Western Australia. Journal of Agriculture
Western Australia (2nd Series), 22: 24874.
Durham, L.J. 1958. A Report on a Reconnaissance Survey of Erosion in the Molonglo River
Catchment Area. New South Wales Soil Conservation Service.
Eyles, R.J. 1977a. Changes in Drainage Networks since 1820, Southern Tablelands, NSW.
Australian Geographer, 13: 37786.
Eyles, R.J. 1977b. Erosion and Land Use in the Burra Catchment, Queanbeyan. Journal
of Soil Conservation Service of New South Wales, 33(1): 4759.
Fitzgerald, K. 1968. The Ord River Catchment Regeneration Project, W.A. Dept. Agric.
Bulletin 3599.
Fogarty, P., G. Clifton, B. Griffiths and R. Hammond. 1989. Soil Erosion and Soil Con-
servation in the Lake Burley Griffin Catchment, NSW and ACT. Australian Journal of
Soil and Water Conservation, 2(1): 2832.
Forrest, A. 1880. North West Exploration. Journal of the Expedition from De Gray to Port
Darwin. Perth: Western Australian Government.
271
Human Interaction with Soil-Sediment Systems in Australia
Teakle, L.J.H. 1944. Soil Erosion and its Relationship to Soil Type and Geological Forma-
tions of the Upper Ord Basin. File 23/44 Vol. 1 Western Australian Department of
Lands and Surveys.
Walker, B.H., D. Ludwig, C.S. Holling and R.M. Peterman. 1981. Stability and Semi-Arid
Savanna Grazing Systems. Journal of Ecology, 69: 47398.
Wark, R.J. 1987 Deposition of Sediment in Lake Argyle. Report Number WP47. Western
Australian Water Authority
Wasson, R.J. 1994. Living with the Past: Uses of History for Understanding Landscape
Change and Degradation. Land Degradation and Rehabilitation, 5: 7987.
Wasson, R.J. 2003. Effective Policy Interventions in Environmental Systems using Mate-
rial Budgets. In S.R. Dovers, D.I. Stern and M.D. Young (eds), New Dimensions in
Ecological Economics (Cheltenham: Edward Elgar), pp. 12992.
Wasson, R.J., G. Caitcheon, A.S. Murray, M. McCulloch and J. Quade. 2002. Sourcing
Sediment Using Multiple Tracers in the Catchment of Lake Argyle, Northwestern Aus-
tralia. Environmental Management, 29(5): 6346.
Wasson, R.J. and R.L. Clark. 1985. Environmental History for Explanation and Predic-
tion. Search, 16: 25863.
Wasson, R.J., R.K. Mazari, B. Starr and G. Clifton. 1998. The Recent History of Erosion
and Sedimentation on the Southern Tablelands of Southern-Eastern Australia: Implica-
tions for Soil Conservation. Geomorphology, 24: 291308.
Wasson, R.J., A.S. Murray, H. Crockford and G.G. Caitcheon. 1989. Sources of Sediment
Reaching Lake Burley Griffin. Progress report to ACT Parks and Conservation Service
and NSW Soil Conservation Service.
Wasson, R.J. and A.Yu. Sidorchuk. 2000. History for Soil Conservation and Catchment
Management. In S. Dovers (ed.), Australian Environmental History: Still Settling Australia
(Oxford), pp. 97117.
Weaver, P.M., L.J. Pen and A.E.G. Reed.1994. Modifying the Phosphorus Cycle to Achieve
Management Objectives in the Oyster Harbour Catchment. Water Journal, Feb. 1994:
2832.
Weaver, D.M., P.B. McCafferty and E.G. Reed. 1999. A Phosphorus Inventory for Albany
Hinterland, South Coast of Western Australia. In C. Barber, R. Humphreys and J. Dixon
(eds), Proceedings, International Conference on Diffuse Pollution, Solutions and Innovations
(Perth, May 1999), pp. 10516.
273
Dynamics of Soil, Landscape and Culture on Easter Island
10
A. Introduction
1,200 years. Environments were settled where humans had never set foot before.
Exquisitely adapted land use strategies and techniques were indispensable for the
permanent settlement of human populations on the isolated islands with their poor
resources. Biodiversity of the islands was low because of their extreme isolation.
Therefore domesticated plants were always part of the cargo of the large double-
hull canoes of the Polynesians, as were animals such as poultry, pigs, rats and dogs
for supplementing the fish from the ocean and the meat resources found on the
islands (Orliac and Orliac 1995). The production systems set up on the islands
had to be organised on the principle of subsistence and be sustainably adapted to
their resources if they were to function for a long time.
There are several possible reasons for the unprecedented conquest of the
Pacific region:
curiosity in general and the human urge for conquest and innovation
catastrophes affecting settled regions, e.g. volcanic eruptions, earthquakes,
floods and typhoons
overpopulation of the limited space of the settled regions
shortage of resources due to catastrophes, overpopulation or natural and
man-made changes of the environment.
Settlement took place in three cultural, temporal and spatial phases:
the settlement of western Oceania by seafaring peoples which, coming
from South-east Asia, reached the western part of the South Pacific via New
Guinea and Australia about 35,000 to 40,000 years ago (Kirch 2000).
expansion of the Lapita people, around 3000bc, southward from the
Indonesian-Malayan region, reaching the Papua region by 1500bc and
moving eastward to western Polynesia. Known by its highly developed pot-
tery techniques, the culture group reached the archipelagos of Fiji, Tonga
and Samoa between about 1200 and 1100bc (Orliac and Orliac 1995).
The Lapita culture was the precursor of the Polynesian culture (Esen-Baur
1989).
advance of the Polynesians into the central Pacific region, most likely origi-
nating from the Tonga and Samoa region (Kirch 1984). Around 200bc
the centre of Polynesia was reached. The Hawaiian islands were discovered
about ad400, New Zealand probably around ad1000 (Kirch 2000). Like
all these datings, that of the settlement of Easter Island, the easternmost
island of Polynesia, is still not fully established, with estimates ranging from
ad300 to ad800 (e.g. Orliac and Orliac 1995; Kirch 2000; Lee 2001;
Martinsson-Wallin 2002). The age determinations are based on radiocarbon
276
Andreas Mieth and Hans-Rudolf Bork
Figure 10.1. Stone statues (moai) sticking deeply into the slope sediment of the quarry
volcano Rano Raraku. These moai represent the last phase of the megalithic culture on
Easter Island.
were transported. It is as yet unknown how the statues were lifted from the quarry,
how they were transported downslope, how they were moved along the roads and,
finally, how they could be erected on the platforms. Most archaeologists agree that
tree trunks were indispensable for moving the statues (Esen-Baur 1989, Flenley and
Bahn 2003, van Tilburg 1994, Grau 1998). Depending on the technology assumed,
estimates on the number and quality of trees needed vary considerably. Answering
the question precisely is of considerable ecological significance, though.
About 900 moai are known, of which about 300 were erected on ahu. The
largest statue set up is about 10m high and weighs about 80t. The largest moai
ever produced, however, have remained in the quarry of Rano Raraku, marking
the climax and possibly also the abrupt termination of the moai culture. Impres-
sive testimony to the end of this cultural phase are about 400 statues still standing
or lying on the outer and inner slope of Rano Raraku, many of them buried in
sediment up to their heads, and many never completed nor cut free from the tuff
rock of the volcano.
Why the stone cult grew to such unprecedented dimensions and why it
came to its sudden end are among the largely unanswered questions of the cultural
history of the island. Ecological catastrophes are increasingly believed to have been
the reason for the collapse, but opinions differ as to whether natural or man-made
causes lie behind such a sudden change of environmental and thus also of living
conditions. Flenley and Bahn (2003) created a model describing possible inter-
relationships between land use, ecosystem change and cultural effects.
No one knows how many people lived on Easter Island during the heyday
of the moai culture. Some authors assume a maximum population of 8,00010,000
(Bahn 1993a, van Tilburg 1994, p.52), but there is no reliable basis for any well-
founded estimates. All figures presented are extrapolations from an inferred number
of the original population or have been estimated from the settlement structures
found (Flenley and Bahn 2003).
Around the fifteenth or sixteenth century, a cultural shift occurred,
characterised by but not only by the end of moai production. It possibly came
together with food shortages, increasing social tensions, warfare and a population
decrease to about 2,000 persons (Flenley and Bahn 2003; Bahn 1993a). Evidence of
battles comes from oral traditions, and from weapons produced since the fifteenth
or sixteenth century (Flenley and Bahn 2003). Thousands of obsidian weapons,
such as spearheads, and the flakes from their production litter the surface and are
buried in sediments. Fighting also led to the destruction of the once holy moai.
The statues were toppled from their platforms and their faces, mainly the eyes, were
deliberately destroyed (Flenley and Bahn 2003).
The time of war also saw the rise of new cultural and spiritual ideals expressed
in the so-called birdman cult (Routledge 1919 [1998]; van Tilburg 1994, Martins-
son-Wallin 2002). The spiritual centre of this cultural phase was the ceremonial
279
Dynamics of Soil, Landscape and Culture on Easter Island
village of Orongo on the narrow seaward crater rim of Rano Kau volcano. Ritual-
ised competitions may have served the selection of high-ranking decision-makers
and advisers of the society. Seabirds, new fertility symbols, new gods and new
hierarchic ideals replaced the old peaceful society with its ancestral cult and a life
strictly regulated by taboos. Living in hidden caves became important, probably
as an expression of security needs in a conflict-ridden society.
Perhaps the birdman cult was an effort to transform deadly conflicts into
ritualised fighting. Van Tilburg (1994, pp.5862) sees a close link between changing
factors of the ecosystem, primarily the marine part of it, and the development of the
birdman rituals. Other authors also attribute a shortage of resources and resulting
tensions to the end of the moai culture and the rise of the birdman rites (Flenley and
Bahn 2003; van Tilburg 1994; Martinsson-Wallin 2002). Various datings are given
for the demise of the moai culture: Kirch (1984, quoted in Lee 1992, p.8) thinks
it was around ad1500, for Bahn and Flenley (1992, p.152) it was in the sixteenth
century, for Lee (2001, p.109) around ad1550, and Martinsson-Wallin suggests
about ad1600. Some authors put the end of the moai culture, the beginning of
the birdman culture and/or the beginning of the period of warfare as late as 1680,
based on the seemingly good coincidence between radiocarbon datings from the
western part of Poike Peninsula initiated by Thor Heyerdal and the oral tradition
Figure 10.3. Easter Island with locations of recent investigations on soil history and
landscape development
280
Andreas Mieth and Hans-Rudolf Bork
of a fight having taken place at the same time in that area (e.g. van Tilburg 1994,
p.51; Esen-Baur 1989, p.107). It is assumed that food shortages and warlike
conflicts drastically reduced the population from the seventeenth century onward
(Flenley and Bahn 2003; Bahn 1993a).
On 5 April 1722 the Dutch captain Jacob Roggeveen was the first European
to set foot on Easter Island. There followed a time of increasingly frequent visits
by European seafarers (Gonzales 1770, Cook 1974, La Prouse 1786, v. Kotzebue
1816, Beechey 1825, Palmer 1868, Geiseler 1882, and others). The first European
visitors of the eighteenth century reported moai still standing upright on their ahu,
but by the beginning of the nineteenth century many, and by the end of it most
likely all, of the stone statues had fallen over. This is taken as evidence that fighting
on Easter Island continued well into the nineteenth century.
The seafarers records contain the first population estimates. Gonzales
speaks of 900 to 1,000 persons, Cook of 700 people, most of them men (quoted
by Esen-Baur 1989, pp.56, 59). La Prouse estimated the island population to be
close to 2,000 around 1786, a number Bahn and Flenley (1992, p.179) think to
be plausible. There are considerable uncertainties, though. Most of the European
visitors stayed for a few days only. Roggeveen and Cook saw mostly men on the
island, which may indicate that women and children hid inside their houses or
caves during the ships visits.
In 1866 the first Christian mission was set up on the island. Missionary
activity and slavery in the 1860s brought a rapid end to the rituals of the birdman
cult. The end of this cultural phase can therefore be dated almost by the year. In
1877 the population of Easter Island was reduced to its all-time low of 110 persons
due to deportation of men and imported diseases (McCall 1994, p.64). According
to the 2002 census about 3,800 people live on the island today.
p.9) lists 48 indigenous flowering plants and ferns, eight of them endemic. Most
likely also the autochthonous flora the first settlers met with on Easter Island was
quite poor, but richer than today, judging from archaeological and palaeobotanical
evidence. Due to differences in the long-term resistance of pollen to weathering,
palynological analysis will permit only limited taxonomic differentiation and thus
yield only incomplete information on the earlier diversity of species. From cores
taken from the crater lakes of Easter Island Flenley et al. (1991) palynologically
identified five woody plant species existing prior to the first settlement. They also
identified six types of Compositae-tubiliflorae pollen, but the evidence did not allow
further taxonomic differentiation. In more recent anthracological studies (analysis
of charcoal macro-remains) in several archaeological contexts, nine more woody
plant species could be identified, eight of them taxonomically identified (Orliac
2000; Orliac and Orliac 1998, 2001). Both from pollen and endocarp remains an
indigenous but now extinct palm species was identified (Flenley 1993b). Com-
parison of the fossil with present material suggests that the palm was closely related
to, if not even identical with Jubaea chilensis on the genus or species level (Drans-
field et al. 1984; Grau 2001). Because of the uncertainty, Zizka (1991) classified
this palm tree as a species of a new genus: Paschalococos disperta. Following Grau
(1998, 2001), for the present study the name Jubaea sp. respectively Jubaea-palm
was chosen. Another woody plant that had temporarily become extinct on Easter
Island in the European period, endemic Fabacea Sophora toromiro, now survives
with only a few individuals in some European botanical gardens and in some house
gardens on Easter Island.
In the Polynesian, and more so in the European period, the flora of the
island became strongly modified by anthropochorous (introduced by man) species.
Only 16 per cent of the present flora are indigenous species (Feeser 2003). Perhaps
more than 100 plant species were introduced to the island in the historical period
(Flenley 1993a). Dominant among them, because of the use of the island as graz-
ing land since the end of the nineteenth century, are voluntarily imported grasses,
together with those introduced inadvertently.
It is not easy to reconstruct the prehistoric vegetation from the palaeoeco-
logical evidence. Pollen results only allow limited conclusions as to the frequency
and distribution of the plant species identified. In spite of the extreme isolation of
Easter Island it cannot be fully excluded that pollen in the crater profiles may have
originated from regions outside the island. Similarly charcoal from archaeological
sites yields a selective picture only, reflecting, with a high degree of certainty, the
presence of a species in the area (with the exception of driftwood), but not neces-
sarily its distribution and frequency. Nevertheless some authors have attempted
to describe the island vegetation prior to human settlement. Skottsberg (1953)
assumed that an open park-like forest landscape existed at the time of the first
settlement. Flenley and King (1984), Flenley et al. (1991) and Flenley (1993a,
284
Andreas Mieth and Hans-Rudolf Bork
1993b, 1996, 2001), from their pollen analysis, reconstructed a woodland of palm
trees interspersed with Sophora and other shrubs that covered at least part of the
island. Orliac (2000), from her anthracological studies, reconstructed a mesophytic
forest for Easter Island as it still exists in the lower elevations of other East Polyne-
sian islands. Until now there has been no direct evidence of the distribution and
density of stands of the Jubaea palm. In his archaeological papers, Charles Love
described root traces of this species, mainly from soils of Poike Peninsula (quoted
by Bahn and Flenley 1992, p.90), but did not study them in more detail. From
root tubes, charred endocarp remains and the character of Ah- and B-horizons of
soils Wozniak (1998, pp.1901, 2001, p.93) inferred an open palm-tree vegeta-
tion in the north-west of the island. Our own recent studies on the reconstruction
of a palm-tree dominated vegetation will be described below.
There is consensus among all researchers that there has been a dramatic
change of the island vegetation since the time of settlement, but with many open
questions as to its chronology. This is due to the problems of extracting spatial and
temporal distribution of plant species from pollen and charcoal remains, but also
to problems of dating the pollen stratigraphy obtained from drill cores (Flenley
et al. 1999, pp.923). According to Flenley (1993b, 1996, 1998) and Bahn and
Flenley (1992) the decrease of tree and shrub pollen simultaneously with the in-
Figure 10.4. Grassland and surface covers of volcanic rocks characterise Easter
Islands landscape today.
285
Dynamics of Soil, Landscape and Culture on Easter Island
1
KIA stands for radiocarbon datings carried out in the Leibniz Laboratory for Radiometric
Dating and Isotope Research at the Christian-Albrechts University in Kiel, Germany.
288
Andreas Mieth and Hans-Rudolf Bork
8. Soil Fertility and Soil Erosion on Easter Island: The State of Knowledge
Until recently the soils of Easter Islands were hardly studied. Consequently reli-
able information on former soil fertility or on indications of soil erosion in the
literature is scarce. Studies of the horticulture on Easter Islands have been able to
partly document the prehistoric land-use techniques, but not the earlier soil condi-
tions or soil qualities. No causal analysis of the relationship between soil quality
and land use has therefore yet been possible. Instead, some authors have inferred
former cultivation conditions from the present state of the soils. Presently exposed
volcanic sediments and volcanic rocks, together with the stone-strewn surfaces,
create the impression that large parts of Easter Island carry soils that are infertile
or not suitable for horticulture.
The present high permeability of the volcanic rocks and the almost complete
absence of deep humus-rich soils with a high usable field capacity in large parts
of the island has been extrapolated by a number of archaeologists into prehistoric
times, and has then been thought to have been the main reason for the high techni-
cal input in horticulture in those times. Thus Stevenson and Haoa (1999, p.6), in
their analysis of prehistoric horticultural techniques, assume that soil conditions
in those days would have hardly been different from today. Louwagie and Langohr
(2002), in contrast, emphasise our scant knowledge of the earlier suitability of the
soils for horticulture. The authors took samples from two soil profiles of the island
and applied standard tests to determine their capacity for crop production.
Wozniak (1999) describes the high permeability of the rocks and the re-
duced availability of certain plant nutrients such as nitrogen and potassium as the
limiting factors explaining why prehistoric garden soils were mulched with charred
plant remains. From only a few studies of the present state of the island soils,
conclusions were drawn as to the prehistoric environmental conditions, and these
were described as an uncertain environment for a secure food supply. Spatial and
temporal variability of climate as well as generally less favourable climatic condi-
tions than on other islands of the Polynesian region are regarded as additional risk
factors for prehistoric agriculture (Flenley and Bahn 2003; Hunt and Lipo 2001;
Stevenson et al. 1999, 2002).
As yet there are also no reliable studies on soil erosion in prehistoric times.
The first identification of erosion in prehistoric times was interpreted from cores
taken from the crater lake of Rano Raraku (Bahn and Flenley 1992; Flenley 1993b).
At a depth of 11.2m the authors found a sediment that had been eroded from
the inner slope of the crater. They could only very roughly date it as having been
deposited at some time within the last 1,000 years, and attributed it to quarry-
ing during the principal phase of moai production. From archaeological evidence
Wozniak (1998, pp.189, 191) described a sediment 1m thick from the north-
western coast, at the foot of Maunga Terevaka, which had been deposited before
and during the construction of an ahu. The author attributed it to soil erosion
289
Dynamics of Soil, Landscape and Culture on Easter Island
following forest clearing in the area and dated the deposition to the fourteenth and
fifteenth century. Because of temporal inconsistencies in the distribution pattern of
archaeological surface finds on the upper volcano slopes, she also inferred prehistoric
erosion events, but did not bolster her interpretation with geomorphological and
pedological evidence.
It was not until the middle of the twentieth century that the nature
conservation agencies began to pay attention to young erosion features and the
resulting change of the landscape in large parts of Easter Islands (Diaz Vial 1949).
Allochthonous trees and shrubs were planted, mainly on Poike Peninsula, in order
to protect slopes above areas with severe soil erosion against any further loss of soil.
Later studies showed, however, that many plants had not taken root permanently
and that soil erosion had not been stopped (Sudzuki 1979). Although soil erosion
could easily lead to severe ecological and economical problems on the whole island
because of its isolated position and limited size, there has been practically no sci-
entific analysis of the interrelationship between land use and soil erosion. Neither
temporal and regional differentiation of soil erosion has been undertaken, nor has
there been any calculation of erosion balances.
Figure 10.5. Poike Peninsula is the easternmost and oldest part of Easter Island. Large
areas of severe soil erosion are visible at the eastern tip of the peninsula.
Figure 10.6. The investigation of large soil profiles, here a 100.5 m long exposure above
the south-western cliff of Poike Peninsula, is of significance in the four-dimensional
landscape analysis.
291
Dynamics of Soil, Landscape and Culture on Easter Island
Figure 10.7. Cone-like formation of Jubaea palm root casts in the Holocene soil. The
root casts represent the last palm generations which existed in certain areas.
292
Andreas Mieth and Hans-Rudolf Bork
in cross section. Where the upper part of the root system has been truncated, only
the deeper-seated, more cylindrical part of the tubular pattern has survived. The
diameter of the root discs at the former soil surface corresponds to the diameter of
the palm trunks at their base. Diameters of up to 1.5m were measured.
Generally the root biomass has been completely decomposed, but in some
of the tubes organic relics can be found in the shape of very thin charcoal linings.
More frequently the roots are coated by black to grey sesquioxide coatings (oxides of
iron and manganese). The matrix surrounding the root tubes is very stable. The soil
particles there have been glued together. The polyhedric pattern of the B-horizons
in the root-tube areas is so stable that major aggregates containing root tubes can
be taken from the soil without destroying the structure.
From the distance and diameter of the clearly identifiable root cones of
individual trees within the soil profiles the distribution pattern of the former palm-
tree vegetation could be reconstructed. Along a 100.5m long section in the south-
western part of Poike Peninsula, 29 individual palms of varying size and at varying
distances could be identified (Mieth and Bork 2003a). Based on the morphology of
the present Jubaea chilensis growth heights of the prehistoric palms from 5 to 18m
were reconstructed. The horizontal distances and the inferred height of the palms
allow the reconstruction of a partly dense, partly open palm forest with a storeyed
Figure 10.8. Palm root casts of the extinct Jubaea palm traverse the soil vertically forming
unbranched channels. The channels are frequently coated by iron and manganese oxides
of dark grey colour.
293
Dynamics of Soil, Landscape and Culture on Easter Island
Figure 10.9. True to scale reconstruction of the prehistoric palm vegetation for a slope
segment in the south-west of Poike Peninsula
age structure. The presence, distribution and density of shrubs in this system are
still open to conjecture, as these plants have left no autochthonous traces. The exist-
ence of a prehistoric palm vegetation could be confirmed from several other sites
on Easter Island as well (Mieth et al. 2002; Mieth et al. 2003a; Bork and Mieth
2003). Based on several island-wide test sites we assume that at least 70 per cent of
the island (equal to about 116km) was once covered by the palm forest.
Based on an average growth distance between palms of 2.6m derived
from several sites a former island-wide palm-forest of more than 16million trees
has been calculated. Assuming that there were no efforts to increase the number
of trees after settlement, this would be about the palm population at the time of
initial settlement. A prerequisite for this estimate is that the root traces described
always portray a single, namely the last palm generation at a given site. Had earlier
palms also left their traces, the multiple interference of root systems would no longer
have allowed the identification of individual palms. The situation is the same in all
areas studied. It remains unclear why in each case only the roots of the last palm
generation were preserved.
11. Horticulture in the Palm Forest: The Strategy of Sustainable Land Use
Several stratigraphic traits in the soil profiles of Poike Peninsula confirm that the
palm forest was used for crop production (Mieth and Bork 2003a, Mieth et al.
2003a). Working the soil with a digging stick or appropriate stone tools created
planting pits of different dimensions, depending on the kind of crop planted. Dig-
ging loosened the soil to an average depth of 30cm. Some planting pits were as
much as 60cm deep. Ah- and B-horizons were fully homogenised by the digging.
Because of its mostly crumbly texture, the cultivated soil is clearly different from
the autochthonous relics of the Holocene soil. The base of the cultivated layer is
weakly undulating. Less frequently, at high soil humidity, individual planting pits
can be identified within the cultivated soil. Some of them are deep-reaching, also
294
Andreas Mieth and Hans-Rudolf Bork
affecting the base of the cultivated layer. Occasionally small aggregates of the less
fertile C-horizon have been worked into the lowermost part of the cultivated layer.
In most cases any incorporation of the C-horizon seems to have been avoided.
Most likely the spatial density of planting pits existing at the same time was
low, or planting may have been restricted to smaller areas. Some hundred years of
crop production in the palm forest were sufficient, though, to completely loosen
and homogenise the material of the cultivated layer to a rather uniform depth
between the palms. The available space between the trees was used very efficiently,
but the palm roots were protected, as can be seen from particularly deep planting
pits carefully dug right next to the trunks. The palm roots there were not destroyed
by the digging. The lateral boundaries as well as the (formerly) near-surface parts
of the root cones have been fully preserved in the profiles. The reconstruction of
the vegetation as described above would not have been possible without such a
high degree of root preservation.
The palm forest had important protective functions for the crops. It provided
shade and protected them against transpiration and wind. The forest vegetation
also protected the soil, as the canopy of the trees weakened the impact of rain and
wind and thus prevented soil erosion. The garden soils were sustainably enriched
with organic substance by mulching of plant remains. Therefore the humus con-
tent of the cultivated horizon is higher than that of the original soil. In the oldest
cultivated layers traces of burning are very rare.
In the eastern part of the island the phase of forest-related horticulture
lasted about 600 years, beginning with the first settlement assumed for the seventh
century, and ending in the second half of the thirteenth century with the clearing
of the palm forest.
Figure 10.10. Burned stump of the indigenous Jubaea palm with indications of its
use as umu and palm root channels below the fire site.
Figure 10.11. Burned nuts of Easter Islands extinct palm are found in charcoal layers,
fire sites and umu. These nuts allow the dating of slash and burn events in certain areas
with great exactness.
296
Andreas Mieth and Hans-Rudolf Bork
fireplaces (umu) dug below the burned layer or placed above burned palm
stumps
stacked layers of fine-grained (transported) charcoal particles within fine-
grained sediment immediately above the burned layer
soil baked red at the base of charcoal veins and pockets.
All these findings clearly indicate human forest clearing and burning. From the soil
profiles the systematic and efficient application of the slash-and-burn method can
be reconstructed. The palms were chopped off only a few centimetres above the soil
surface and the leaves and other useless parts were probably removed at the place of
felling. The trunks were then taken out and put to various uses. The remains of the
palms, their stumps, the litter including some fallen-down nuts as well as the
waste from other non-usable plants were burned. The traces of gnawing by rats on
the nuts (Mieth et al. 2002) are evidence that they had already lain at the surface
before the fire. Sometimes dry grass, twigs and leaf litter were heaped directly on
the palm stumps, perhaps to speed up incineration. Burned layers extending over
hundreds of square metres, e.g. at several places on Poike Peninsula, confirm that
large tracts of land were burned at a time. The preserved thickness of the surviving
autochthonous charcoal layers up to several millimetres as well as the frequently
stacked deposition of finely distributed charcoal at the base of bedded sediments
suggest a considerable local thickness of the burned layer right after the fire, in view
of the low stability of charcoal dust and ashes under the influence of rain and wind.
Similarly the local, heat-induced chemical alteration of the clay soils turned red
by fire speaks of a considerable intensity of the fires.
Radiocarbon datings confirm that burning the palm forest was not a single
event, but consisted of a sequence of clearings, each limited in space. Datings from
Poike Peninsula and a site on the Rano Raraku volcano allow the reconstruction
of the consecutive phases of forest clearing by fire:
The radiocarbon dating of a palm nut from a burned layer in the easternmost
part of Poike Peninsula of 731+25bp (2- cal ad12441254, 12561299,
KIA 17107) puts the beginning of fire clearing into the second half of the
thirteenth century (Mieth et al. 2002). Two datings of charcoal from a
large fireplace in the same area correlate quite well (631+22bp, 2- cal
ad12971329, 13431396, KIA 18837; 588+22bp, 2- cal ad13041367,
13841408, KIA 18838).
The youngest dating of a palm relict originates from the east of Poike Pe-
ninsula, only a few hundred metres away from the site just referred to. A
burned nutshell from the fireplace there yielded a radiocarbon dating of
317+20bp (2- cal ad14921505, 15071600, 16141642, KIA 20383).
This does not stand for a specific clearing at a specific place, though, but
297
Dynamics of Soil, Landscape and Culture on Easter Island
In the east of Poike occasional planting pits dug down from the burned layer suggest
occasional planting right after the clearing (Mieth et al. 2002).
In the south-west of Poike Peninsula development of a humus horizon
indicates a longer phase of calm after the clearing of the forest. Within some dec-
ades or even a few centuries a pronounced humic horizon developed underneath
a grassland vegetation. Evidence of horticultural activity at this site is 100 to 200
years younger than the end of the first land-use phase, respectively the clearing of
the forest. This can be read from the following characteristics found in a 110.5m
long soil section above the cliff (Mieth and Bork 2003a, Mieth et al. 2003a):
Starting from several centimetres of a dark, grey-brown humus horizon there
are clearly delineated planting pits, 10 to 50cm deep. Their U-shaped cross
sections are up to 40cm wide and 10 to 30cm apart.
Between the planting pits there are undisturbed relics of the Ah-horizon
which also cut through the upper parts of palm root relics.
In contrast to the older cultivated layer the base of the younger one is quite
indented because of the varying depth of individual planting pits.
The root relics of the palms have only occasionally been affected by the
digging.
The planting pits contain loose fillings of the dark grey-brown material of
the humus horizon, frequently mixed with charcoal and ash.
The dating of charcoal from the younger cultivated layer yielded a radiocar-
bon age of 319+21bp (2- cal ad14921600, 16141641, KIA 18834)
The well-defined planting pits, the relics of the Ah-horizon between them as well
as the good preservation of the palm root relics all indicate that also in the south-
western part of the peninsula there was a short episode of at best a few years or
decades duration of low-intensity horticulture after the palm-forest clearing.
Figure 10.12. Layered fine-grained sediment: result of rising sheet erosion after woodland
clearance and intensification of land use on Poike Peninsula. Single layers of this colluvium
represent single erosion and deposition events.
Only the uppermost parts of the volcano slopes, close to the divide, did not
receive any of the finely bedded deposits. On the young micropediments and in
the gullies of Poike Peninsula they have been removed by later erosion. The profiles
studied show an average deposition of 60, in certain cases of up to 400 thin layers
(Mieth et al. 2003). Generally two layers, in most cases just a few millimetres thick,
form a unit that is representative of a single rainfall and runoff event. The upper layer
of each unit consists of clayey micro-aggregates of silt to fine sand size, the lower
layer of microaggregates and grains of mostly sand and grus size. The dominant
clay minerals are kaolinite and iron oxides. Flow structures, grain-size distribution,
bedding directions and the general structure of the deposits are evidence of fluvial
erosion and deposition.
Analysis of downslope catenas revealed a kind of imbricated bedding of se-
quences about 10m long and 70 to 150cm thick, each with the fine-layer structure
described above. From their position it is clear that the sediment packages deposited
on the lower slope are older than those in mid-slope and upper-slope position, the
oldest colluvia being those at the bottom of the lowermost sequence. The vertical
profiles of the older and younger deposits have clearly distinct sedimentary and
300
Andreas Mieth and Hans-Rudolf Bork
Figure 10.13. Prehistoric phases of land use and soil degradation, reconstructed for a
downslope segment above the cliff in south-west Poike.
A. Agroforestry in a palm tree forest until about ad 1450
B. Slash and burn of the palm forest around ad 1450
C. Development of a humic horizon from about ad 1450
D. Few planting pits result from extensive agriculture from about ad 1550
E. Deposition of a layered fine-grained sediment after about ad 1675
layers. Some of the sediment will have been washed over the cliff; thus the erosion
rates mentioned below are minimum estimates. Because of the gently concave lower
slopes above the cliff, where the fines could easily be trapped, sediment loss into
the sea will have been well below 20 per cent. Approximately 640 years after its
beginning the erosion reached the near-summit slope areas of Pua Katiki volcano.
Agriculture on the eroded land became impossible.
In contrast, intensive land use well into the last three centuries is reflected
in the colluvium on the slopes of Maunga Orito in the south-west of Easter Is-
land. There numerous embedded obsidian tools, weapons and charcoal as well as
planting pits and umu are evidence of an intensive recent phase of land use (2
cal ad16651683, 17331784, 17891808, 19281955, KIA 17116; 2 cal
ad16221699, 17231779, 17981814, 18321879, 19151944, KIA 17117;
2 cal ad16821734, 18071903, 19051930, KIA 17118).
302
Andreas Mieth and Hans-Rudolf Bork
vegetation damage and soil compaction along animal tracks triggered linear erosion.
Originally track-wide rills rapidly expanded. The gullies today are up to 6m deep, up
to 50m wide and several hundreds of metres long. Lateral erosion and the removal
of walls between them created extended micropediments on their floors.
The formation of gullies and micropediments has thus been the outcome of
barely four decades of sheep-grazing (from the 1930s to the 1960s). Comparison
of todays situation with aerial photographs from 1981 shows that there has not
been much change of the erosion forms since then. The slowing of erosion after
the early 1960s is probably due to the increasing vegetation cover after the end of
grazing and annual grass burning on Poike Peninsula. The numerous cattle tracks
with their compacted soils, though, still cause local runoff concentration and linear
erosion. Check dams made of metal plates from cut-up oil drums set up on the
gully heads of eastern Poike had at best no effect at all, or even accelerated gullying
there (Mieth and Bork 2003b).
cultivated a range of plants without any overall forest clearing. The positive effects
of the forest vegetation were used and preserved for centuries. The kaolinite- and
sesquioxide-rich topsoil between the trees of the palm forest was mechanically
loosened up and homogenised by cultivation. Mulching of plant litter greatly
increased the humus content of the cultivated layer and thus led to high soil fertil-
ity. The complete preservation of the tube systems of the palm roots in the areas
of deposition is proof that mechanical soil treatment was careful. In this phase of
sustainable horticulture erosion was avoided. The integration of various garden
crops with the existing palm forest vegetation and only low and selective use of
the trees themselves corresponds to the type of sustainable agroforestry described
in the South Pacific by Hunter-Anderson (1998) and Kirch (2000).
It is not yet known whether there was any deliberate tree cultivation in those
first centuries. As seeding and planting of Cocos nucifera and other trees (e.g. Arto-
carpus sp.) are common on other South Pacific islands, described as arboriculture
by Kirch (2000), it is quite likely that this was also the case on Easter Island. It is
also a matter of speculation as to what extent the palm trees were used. It is likely
that trunks and leaves were used for a range of purposes (Grau 1998; Orliac 2003).
Some parts may also have been used for human consumption. Grau (1989), for
instance, thinks that the hearts of palm saplings may have been used for food.
Eating the pulp of Jubaea nuts similar in taste to the endosperm of Cocos nucifera
is still mentioned in the reports of the European discoverers. Cracked nutshells in
cooking and refuse pits equally indicate that the palm fruits were consumed (Orliac
2003). Grau (1998, p.122) writes that also the soft exocarp of the fruits is edible.
In view of an average annual yield of 100kg of nuts from a fully grown palm tree
(Grau 1989) and the calculated figure of 16million palms on the island this must
have been an enormous food potential. The same is true for the sugar-rich juice
of the palms, which on the South American continent is still collected from felled
trunks of Jubaea chilensis and processed into palm honey (Darwin 1999). There is
as yet no evidence of it on Easter Island, but it is quite unlikely that such a valuable
resource should have been overlooked for centuries. Even if the maximum yields of
400litres per trunk common in Chile today (Grau 1998; Orliac 2003) were not
reached on the island, simply because of the large number of trees, this should have
been another abundant food resource (Bork and Mieth 2003). Food and non-food
uses of the palm would not have been mutually exclusive.
From these data and inferences it appears that a reassessment of the Jubaea
palm should be made in view of its economic function during the early culture
phase of Easter Island. It seems that the early land-use phase was one of cold
processing of material and food resources. In the oldest anthropogenic layers of
all the profiles studied there are no umu, perhaps an indication that plants and
animals were eaten raw then. The use of burned plant remains as fertiliser seems
not to have been significant, in contrast to the later horticultural phases. Even tiny
305
Dynamics of Soil, Landscape and Culture on Easter Island
fragments of charcoal are hardly ever found in the oldest cultivated layers, which
makes it difficult to date the beginning of the early phase of horticulture exactly.
Its end, however, because of the disappearance of the palm forest, shows up very
well in the stratigraphies.
creation of garden plots is unlikely. Instead, she postulates the former existence of
a soil- and plant-preserving agroforestry, which, from her conclusions, would have
continued if the forest had not been destroyed by extreme climatic events. In her
view the pollen studies by Flenley show that there had already been a retreat of
the forest before the first settlement (Hunter-Anderson 1998, p.89). Any negative
effects the Polynesian rat might have had on the reproduction of palms she thinks
unlikely; quite to the contrary, opening up of the exocarp may even have helped
germination of the nuts.
Orliac and Orliac (1998, 2001) and Orliac (2000, 2003), from their an-
thracological studies, report a significant decrease of charcoal from woody plants
in favour of burned grass in numerous fireplaces and refuse pits of the seventeenth
century. The authors take this as evidence of an abrupt change from a forest- to a
grass-dominated vegetation at that time. They caution the reader, however, that
the frequency and composition of charcoal in archaeological structures is evidence
that the species existed at a given time, but will yield no information on the fre-
quency and distribution of the plant species. In their conclusions the authors also
tend towards a climatically induced change of the ecosystem. They note a rapid
undifferentiated retreat of the forest. Because of the climatically induced dying of
the forest there would have been an oversupply of firewood for several decades,
before a lack of firewood made grass a substitute (Orliac and Orliac 1998, p.132).
The hypothesis is contradicted by Orliac (2003), though. The youngest palm rel-
ics (charred endocarpon) dated are from the middle of the fifteenth century, and
among several thousand pieces of charcoal from the seventeenth century she found
merely four pieces of Jubaea.
The latest research results, presented in this chapter, are unequivocal evi-
dence of the anthropogenic destruction of the forest of Easter Island. The in situ
relics of palm stumps and nuts within the burned layer, together with the absence
of any larger pieces of burned wood in the soil profiles of Easter Island, are proof
of slash-and-burn forest clearing, where the trunks were extracted for efficient
use and the unusable remains were just as efficiently burned. The root tubes in
the soil are circumstantial evidence of human forest clearing by fire. As described
above, only the root tubes of the last palm-tree generation have been preserved
by agglutination of the soil particles surrounding them just the generation that
was destroyed by fire. Thus only forest clearing followed by fire can have preserved
them. In the central-Chile national park of La Campana a region where Jubaea
chilensis used to be cultivated no such open root tubes could be found (Bork
unpublished, 2003), which leaves fire as the only explanation for the agglutination
of the soil particles. Direct heat has to be excluded, however, as the root tubes have
been preserved down to a depth of several metres. A possible explanation might be
physiological reactions of the fresh palm stumps during a fire by way of excretion
from the roots, but there is clearly a need for more research.
307
Dynamics of Soil, Landscape and Culture on Easter Island
The spatial and temporal retreat of the palm forest, in particular in the east
of Easter Island, can be described in much detail from the stratigraphic studies
undertaken. It appears that deforestation of once completely palm-covered Poike
Peninsula took more than 200 years, from about ad1250 to about ad1450. There
was no reforestation on the peninsula after forest clearing. The local regeneration
of palm vegetation after a first phase of clearing on Rano Raraku is an exception.
As dated, the newly grown palms lasted there for no more than 150 years. There
is no evidence as to whether the stand resulted from spontaneous regrowth or was
due to afforestation.
Our own radiocarbon datings of Jubaea relics are in good temporal agree-
ment with the Jubaea- (Paschalococos) datings by Orliac (2003, pp.1956). Most of
her datings of wood and nut fragments from archaeological structures fall between
about ad1200 and ad1450. According to her there is almost no evidence of
younger palms, which is confirmed by our own stratigraphic analyses. The dating
of a palm nut from a fire site on Poike from the sixteenth to seventeenth century
should rather be regarded as an exception and as evidence that a few specimens of
the species could survive the forest clearing for a few more centuries, but not so
the palm woodland as a whole.
Our own studies as well as those by Orliac (2003) have not yielded any
Jubaea charcoal from the time before ad1250. Neither our own stratigraphic stud-
ies nor Orliacs studies (2000) answer the question as to whether forest clearing
in other parts of the island really began as early as ad8001000, as postulated by
Flenley (1993b) from his palynological studies and only a very few radiocarbon
datings. These datings are not only earlier than the main phase of clearing proved
for Poike and inferred for the rest of the island; they are also close to the assumed
beginning of settlement. As we have calculated, clearing of the island forest within
an assumed period of 400 years would have taken an average of 500 workers for
cutting and processing of the palms (Mieth et al. 2003a). Such a large workforce
would only have been available a few centuries after the initial settlement. Therefore
only spatially limited clearing is likely to have taken place before ad1200.
The stratigraphic evidence of extended burning and the changes in the
soils following them unequivocally show that, together with the areal clearing of
the palm forest, also the extent of the shrub component of the mesophytic forest
was much reduced. The anthracological evidence by Orliac (2000) of a still highly
diverse composition of wood species in the fire sites of the seventeenth century
should not be interpreted as evidence of a forest that was still species-rich. More
likely branches were laboriously collected from small relict woodlots, which became
fully exhausted by the middle of the seventeenth century so that grass had to be
collected for making fire.
There may be a number of reasons why the forest was cleared. The studies
in the east, north and south-west of Poike Peninsula show that immediately after
308
Andreas Mieth and Hans-Rudolf Bork
clearing the areas were used for new settlements and as ceremonial and burial plots
(Mieth et al. 2002; Mieth and Bork 2003a). No doubt population growth created
a demand for cleared land and was the reason why new settlements sprang up in
even the most remote parts of the island. The period of forest clearing on Poike
Peninsula between ad1250 and ad1450 (possibly with some stands remaining
until about ad1600) is in good agreement with the time of increased expansion
of settlement to areas higher up and farther away from the coast of Easter Island as
identified by Stevenson et al. (2002) and Martinsson and Wallin (2002).
The intensification of horticulture specifically the rapid transition from
agroforestry to horticulture in the open was possibly another central reason for
clearing the forest. It is true that the soil profiles from lower slope positions in the
east and south-west of Poike Peninsula suggest that there were only weak attempts
at horticulture after forest clearing. The colluvia overlying the cultivated layers as
well as occasional post-clearing planting pits at the base of the colluvium indicate,
however, that at least for some time after the clearing intensive disturbance on the
upper slopes must have existed, triggering the upslope-progressing sheet erosion and
degradation of the soils. Keeping the areas open by horticulture and thus prone
to erosion is one of a limited number of plausible reasons for the disturbances.
With increasing population an increased demand for timber is another likely
reason for the gradual deforestation of the island. Extraction of timber was certainly
more important than that of firewood, as can be deduced from the almost complete
absence of burned pieces of trunks in the soil profiles and the archaeological struc-
tures (Orliac 2003, p.191). Extracting palm juice is another possible explanation
for systematic forest clearing, as the trees had to be felled to get the sugary juice
(Bork and Mieth 2003). The wood could be put to other uses afterwards.
It is still not clear, however, how most of the trunks were used. Given the
estimated number of 16 million palms before clearing began, their exclusive use
for the construction of houses and boats is highly unlikely. Whether the transport
of the stone statues called for large amounts of trunks may only be answered after
more research into the transport techniques. Assuming that most of the moai
transported from the quarry of Rano Raraku were produced within 300 years, not
more than one or two statues would have had to be transported per year. Even if a
few hundred trunks were used up, not more than at best a few thousand of them
would have been used in a year, and fewer than one million for the total produc-
tion period of the statues.
20. The Late Phase of Prehistoric Land Use: Interrelationships between Landscape
and Cultural Development
In all probability the changes in the landscape and the soils had not only ecologi-
cal, but also cultural consequences. Some considerations on such interrelationships
309
Dynamics of Soil, Landscape and Culture on Easter Island
will be presented below. To a large extent they will have to be speculations, and in
some cases, only questions.
With the clearing of the palm forest of Easter Island the principle of a sustain-
able, soil-protecting subsistence economy was abandoned. A high turnover of the
natural resources became the characteristic of the new type of land use. With high
labour input and obviously also technical efficiency the forest was cleared within a
few centuries. The cold use of the resources was replaced by hot processes. Fire
sites and umu in large numbers dot the soil horizons and sediments after forest
clearing. Some of them, installed immediately after the clearing, had sizes unknown
from earlier times. In the east of Poike Peninsula a box-shaped umu 2.7m wide
and 1.7m deep was found and dated to that time. Several thick layers of charcoal
and numerous animal and vegetable food remains are evidence of several meals
prepared there. The tradition of cooking food in such sunken ovens was continued
well beyond the end of the last wood resources. A large number of umu occur in
the colluvia of the last three centuries, as on the lower slope of Maunga Orito. For
some time firewood could still be collected from isolated stands of woodland, but
later had to be replaced by low-grade fuels.
Figure 10.14. Uncommon large prehistoric umu (fire place for cooking) in eastern
Poike, constructed after woodland clearance. The trench-like feature is filled by several
thick layers of charcoal, ashes and red burned clay and contains a broad variety of food
relics (e.g. bones of chicken and rats, shells of various molluscs and sea-urchins, coral
fragments, palm nuts).
310
Andreas Mieth and Hans-Rudolf Bork
Figure 10.15. Gully in the north-east of Poike Peninsula, a result of overgrazing by sheep
and frequent burning of the grassland in the first half of the twentieth century
Figure 10.16. Badlands in eastern Poike with volcanic bedrock exposed by severe soil
erosion. Basaltic stones on the surface do not originate from this area, they are brought
by man to this area and are relics of prehistoric settlements and cultural places which
have been destroyed by erosion.
312
Andreas Mieth and Hans-Rudolf Bork
because of the absence of any protective tree cover, the surfaces were fully exposed
to the extreme climatic conditions.
Cropland in the centre, in the south, north and west of the island now be-
came protected by two extremely labour-intensive horticultural techniques new to
Easter Island: stone mulching on the surface (Bork et al. 2004) and the construction
of manavai, little gardens surrounded with windbreaks of stone (Stevenson et al.
2002). The stones spread out on the soil surface were probably supposed to protect
against transpiration and wind, dampen the temperature curve at the surface and
give mechanical support to plant growth. Perhaps the gardeners of that time may
also have applied these measures to reduce runoff, increase infiltration and reduce
soil erosion. A recent island-wide analysis of stone mulching and the amount of
work involved has shown that in the late prehistoric and perhaps also in the earli-
est historic phase more than a billion stones (mainly basalts of Maunga Terevaka)
were taken from numerous small quarries and spread out on the garden land of
Easter Island (Bork et al. 2004). The total area with an anthropogenic stone cover
was about 76km, with an average of 150,000 stones per hectare. A calculation
of the amount of work necessary for stone mulching showed that more than 100
persons may have been necessary for about 400 years for cutting, transporting and
distributing the stones (Bork et al. 2004). The third and last prehistorical phase of
horticulture thus obviously was very labour-intensive.
At least from the fifteenth to sixteenth century onward, the clearing of the
palm forest, because of the ecological and land-use change involved, had consider-
able detrimental effects on the population of Easter Island. The following ecological
and economic consequences can either be reliably derived or, in some cases, at least
inferred from the stratigraphic analysis:
loss of the Jubaea palm and its usable parts (wood, nuts, palm juice, palm
hearts)
onset of extreme climatic conditions in the immediate environment of
the people, leading to more difficult and insecure site conditions for the
cultivation of plants at the remaining garden plots
destruction of crops, houses and ceremonial places by frequent runoff
events
high competition for land, specifically over-use of the remaining horticultural
sites
problems of food supply due to the simultaneous loss of good soils, loss
of nutrient-supplying species of the natural vegetation (mainly the Jubaea
palms) and to decreasing yield (in part also because of the climatically
unfavourable exposure of the cultivated plants), and at the same time
313
Dynamics of Soil, Landscape and Culture on Easter Island
high labour input and strenuous work for the stone-based horticultural
techniques.
The factors identified in this recent study confirm and complete the theoretical
model of the interrelationships of ecosystem, land-use and cultural changes on
Easter Island as proposed by Bahn and Flenley (1992). Our studies indicate that
there was already a dramatic change of economic and ecological conditions in the
labour-intensive megalithic culture in the fifteenth or at the latest at the beginning
of the sixteenth century. This is in contrast to other authors who place the collapse
of the system in the seventeenth century or, without substantial proof, even exactly
in the year 1680 (Esen-Baur 1989; Van Tilburg 1994).
The soil degradation which followed the clearing of the forest had a de-
cisive influence on the ecological and economic situation. The key effect of soil
change has so far either been overlooked or underestimated. As yet it can only be
speculated, though, what role the changing vegetation and soils played in the col-
lapse of the moai culture, or whether it triggered the rise of the birdman cult. The
coincidence in time itself between the end of forest clearing and the final phase
of moai production at the turn of the fifteenth to the sixteenth century suggests
a causal relationship between ecosystem and cultural change or even disruption.
Whether it was the scarcity of tree trunks for transporting the giant stone figures,
the high labour input necessary for the stone-mulching system, increasingly dif-
ficult crop production and failed harvests because of lowered soil fertility and the
scarcity of soils suitable for horticulture, or whether there were social tensions and
diseases resulting from deteriorating nutrition that led to the collapse of the unique
megalithic culture, will most likely never be found out. However, the new results
presented here fully support the hypothesis that it was mainly the anthropogenic
and not any natural change of the environment that profoundly altered the cultural
and social situation on Easter Island.
By the example of the Poike Peninsula it has been shown that the almost
ubiquitous soil erosion triggered by forest clearing by the end of the thirteenth
century and kept alive by the types of land use following it continued for more than
600 years and only came to an end when the intensive phase of European land use
had already begun. A single and perhaps only brief disturbance of the surface may
have been necessary to trigger erosion, possibly the one-time forest clearing itself.
Once begun, however, the erosion continued as a dynamic process and only came
to an end when the erosion scarp had almost reached the crater rim of the highest
volcanic structure of the peninsula. Even long phases without any land use, after
settled and horticultural areas had been abandoned from the fourteenth century
onward, could not stop the erosion process. It even continued throughout the last
centuries before European discovery, when the land-use intensity had been much
reduced or had been restricted to the coastal areas.
By the beginning of the twentieth century sheet erosion was replaced by
linear erosion triggered by European sheep raising. This phase of land use also
had long-lasting effects. A few decades of intensive grazing were enough to start
a degradational dynamism that is still active today. So far none of the measures
of conservation management have succeeded in checking it. The first European
phase of land use also had a long-lasting effect on the vegetation. On one hand it
was the grazing and trampling by sheep that completely destroyed the remaining
indigenous and endemic flora, and with it the final chance for regeneration of the
typical vegetation of the island. At the same time, right up to the end of the twen-
tieth century, there has been an introduction of allochthonous species for pasture
improvement and forestry, but also for reasons of nature protection, leading to a
massive anthropogenic transformation of the island flora (Feeser 2003). It also
should not be overlooked that the existing prehistoric cultural remains are severely
endangered by the rapid linear soil erosion.
have existed in Polynesian times on Easter Island or still present in other parts of
southern Polynesia. First trials with Jubaea chilensis on Easter Island have shown
that it grows much better there than on the Chilean mainland. This should be taken
as encouragement for large-scale planting of this tree that was so characteristic of
the early prehistoric environment. Planting shrubs in the terrain already affected
by severe erosion, as tried in the 1970s in the east of Poike Peninsula, will certainly
be unsuccessful. Continuing runoff and erosion dynamics, together with a nutrient
deficit, high resistance of the little-weathered country-rock to root penetration and
poor water supply there, as well as the considerable exposure to solar radiation and
wind in those areas would kill all the plants in a short time.
they hope that this undemanding plant might still be sufficiently productive with
increasing nutrient deficiencies? Was the extreme flourishing of the moai culture
at the time of forest clearing, as well as the enormous growth of the settled areas
and number of ceremonial sites perhaps also due to new influences from abroad?
Most likely, some of these questions will remain unanswered forever.
A particular research effort should be devoted to the extinct palm of Easter
Island. Soil ecological studies should determine how the root tubes of the last palm
generation could be preserved in the soils. Also the significance of the palm as
possibly the most important material and food resource of the early Easter Island
culture deserves more scientific attention, in particular as it seems to have been
unique in the Polynesian cultural realm.
References
Bahn, P. 1993a. The History of Settlement on Rapanui. In S. R. Fischer (ed.), Easter Island
Studies: Contributions to the History of Rapa Nui in Memory of William T. Mulloy, Oxbow
Monograph 32 (Oxford: Oxbow Books), pp.535.
Bahn, P. 1993b. A Brief History of Rapa Nui Archaeology up to 1956. In S.R. Fischer (ed.),
Easter Island Studies: Contributions to the History of Rapa Nui in Memory of William T.
Mulloy, Oxbow Monograph 32 (Oxford: Oxbow Books), pp.738.
Bahn, P. and J. Flenley. 1992. Easter Island Earth Island. London: Thames and Hudson
Ltd.
Baker, P.E. 1997. Petrological Factors Influencing the Utilization of Stone on Easter Island.
In C.M. Stevenson, G. Lee and F.J. Morin (eds), Easter Island in Pacific Context, South
Seas Symposium, Proceedings of the Fourth International Conference on Easter Island
and East Polynesia (Los Osos: Easter Island Foundation), pp.27983.
Bork, H.-R., H. Bork, C. Dalchow, B. Faust, H.-P. Piorr and T. Schatz. 1998. Landschaft-
sentwicklung in Mitteleuropa. Gotha/Stuttgart: Klett-Perthes
Bork, H.-R., H.R. Beckedahl, C. Dahlke, K. Geldmacher, A. Mieth and Y. Li. 2003. Die
erdweite Explosion der Bodenerosionsraten im 20. Jh.: Das globale Bodenerosionsdrama
geht unsere Ernhrungsgrundlage verloren? Petermanns Geographische Mitteilungen,
147: 1625.
Bork, H.-R. and A. Mieth. 2003. The Key Role of Jubaea Palm Trees in the History of Rapa
Nui: A Provocative Interpretation. Rapa Nui Journal, 17(2): 11922.
Bork, H.-R., A. Mieth and B. Tschochner. 2004. Nothing but Stones? A Review of the
Extent and Technical Efforts of Prehistoric Stone Mulching on Rapa Nui. Rapa Nui
Journal, 18(1): 1014.
Coblentz, B.E. 1978. The effects of Feral Goats (Capra hircus) on Island Ecosystems.
Biological Conservation, 13: 27986.
Darwin, C. 1999 (reprint). The Voyage of the Beagle. Amherst: Prometheus Books
Diamond, J. 2005. Collapse, How Societies Choose to Fail or Succeed. New York: Viking,
Penguin Group.
317
Dynamics of Soil, Landscape and Culture on Easter Island
Diaz Vial, C. 1949. Asociacin de Suelos de Isla de Pascua, Apertado de Agricultura Tcnica
IX (Santiago de Chile: Ministerio de Agricultura), pp.11534.
Direccin Meteorolgica de Chile, Departamento Climatologa y Meteoroloa Aplicada. 2002.
Normales de temperatura media mensual (C), periodo 19611990 (unpublished).
Dransfield, J., J.R. Flenley, S.M. King, D.D. Harkness and S. Rapu. 1984. A Recently
Extinct Palm from Easter Island. Nature, 312: 7502.
Esen-Baur, H.-M. 1989. Megalithische Steinplastiken und monumentale Architektur. In
H.-M. Esen-Baur (ed.), 1500 Jahre Kultur der Osterinsel (Mainz: Verlag Philipp von
Zabern), pp.87108.
Etienne, M., G. Michea and E. Diaz. 1982. Flora, Vegetacion y Potencial Pastoral de Isla de
Pascua. Santiago de Chile: Faculdad de Ciencias Agrarias, Veterinarias y Forestales.
Feeser, I. 2003. Untersuchungen zum Grnland der Osterinsel (Chile). Diplomarbeit am
Botanischen Institut der Mathematisch-Naturwissenschaftlichen Fakultt der Christian-
Albrechts-Universitt zu Kiel (unpublished).
Fischer, H. 1999. Schatten auf der Osterinsel, Pldoyer fr ein vergessenes Volk. Oldenburg:
Bibliotheks- und Informationssystem der Universitt Oldenburg
Fischer, S.R. and C. Love 1993. Rapanui: The Geological Parameters. In S.R. Fischer (ed.),
Easter Island Studies. Contributions to the History of Rapa Nui in Memory of William T.
Mulloy, Oxbow Monograph 32 (Oxford: Oxbow Books), pp.16.
Flenley, J.R. 1993a. The Present Flora of Easter Island and its Origins. In S.R. Fischer (ed.),
Easter Island Studies. Contributions to the History of Rapa Nui in Memory of William T.
Mulloy, Oxbow Monograph 32 (Oxford: Oxbow Books), pp.715.
Flenley, J.R. 1993b. The Palaeoecology of Easter Island, and its Ecological Disaster. In S.R.
Fischer (ed.), Easter Island Studies. Contributions to the History of Rapa Nui in Memory of
William T. Mulloy, Oxbow Monograph 32 (Oxford: Oxbow Books), pp.2745.
Flenley, J.R. 1996. Further Evidence of Vegetational Change on Easter Island. South Pacific
Studies, 16(2): 13541.
Flenley, J.R. 1998. New Data and New Thoughts about Rapa Nui. In C.M. Stevenson,
G. Lee and F.J. Morin (eds), Easter Island in Pacific Context, South Seas Symposium,
Proceedings of the Fourth International Conference on Easter Island and East Polynesia
(Los Osos CA: Easter Island Foundation), pp.1258.
Flenley, J.R. 2001. Forest and Civilization on Easter Island. In Y. Yasuda (ed.), Forest and
Civilization (New Delhi), pp.5562.
Flenley, J.R. and S.M. King. 1984. Late Quaternary Pollen Records from Easter Island.
Nature, 307: 4750.
Flenley, J.R. and P. Bahn. 2003. The Enigmas of Easter Island. Oxford/New York: Oxford
University Press.
Flenley, J.R., J.T. Teller, M.E. Prentice, J. Jackson and C. Chew. 1991. The Late Quater-
nary Vegetational and Climatic History of Easter Island. Journal of Quaternary Science,
6(2): 85115.
Forster, G. 1784, reprinted 1983. Reise um die Welt. Frankfurt am Main: Insel Verlag.
318
Andreas Mieth and Hans-Rudolf Bork
Fosberg, F.R. 1963a. The Island Ecosystem. In F.R. Fosberg (ed.), Mans Place in the Island
Ecosystem: A Symposium (Honolulu: Bishop Museum Press), pp.17.
Fosberg, F.R. 1963b. Disturbance in Island Ecosystems. In J.L. Gressitt (ed.), Pacific Basin
Biogeography (Honolulu: Bishop Museum Press), pp.55761.
Genz, J. and T.L. Hunt. 2003. El Nio/Southern Oscillation and Rapa Nui Prehistory.
Rapa Nui Journal, 17(1): 714.
Gill, G.W. 1993. William Mulloy and Rapanui. In S.R. Fischer (ed.), Easter Island Stud-
ies. Contributions to the History of Rapa Nui in Memory of William T. Mulloy, Oxbow
Monograph 32 (Oxford: Oxbow Books), pp.XI-XII
Grau, J. 1998. The Jubaea Palm, Key in the Transportation of Moai on Easter Island. In
C.M. Stevenson, G. Lee and F.J. Morin (eds), Easter Island in Pacific Context, South Seas
Symposium, Proceedings of the Fourth International Conference on Easter Island and
East Polynesia (Los Osos CA: Easter Island Foundation), pp.12024.
Grau, J. 2001. More about Jubaea chilensis on Easter Island. In C.M. Stevenson, G. Lee
and F.J. Morin (eds), Pacific 2000, Proceedings of the Fifth International Conference on
Easter Island and the Pacific (Los Osos CA: Easter Island Foundation), pp.8790.
Heyerdahl, T. and E.N. Ferdon (eds). 1961. Reports of the Norwegian archaeological expedi-
tion to Easter Island and the East Pacific, Vol 1: Archaeology of Easter Island, Monographs
of the School of American Research and the Museum of New Mexico. London: Allen
and Unwin.
Honorato, R., A. Norero, J. Rodriguez and D. Suarez. 1991. Ambiente Agrofsico y Fer-
tilidad de Suelos. In H. Benedetti (ed.), Estudio de la Productividad Silvoagropecuaria
del Fundo Vaitea, Isla de Pascua (Santiago de Chile: Pontificia Universidad Catlica de
Chile, Facultad de Agronoma), pp.2543.
Hunt, T. and C.P. Lipo. 2001. Cultural Elaboration and Environmental Uncertainity in
Polynesia. In C.M. Stevenson, G. Lee and F.J. Morin (eds), Pacific 2000, Proceedings
of the Fifth International Conference on Easter Island and the Pacific (Los Osos CA:
Easter Island Foundation), pp.10315.
Hunter-Andersson, R. 1998. Human vs Climatic Impacts at Rapa Nui: Did the People
Really Cut Down All those Trees? In C.M. Stevenson, G. Lee and F.J. Morin (eds),
Easter Island in Pacific Context, South Seas Symposium, Proceedings of the Fourth
International Conference on Easter Island and East Polynesia (Los Osos CA: Easter
Island Foundation), pp.8599.
Kirch, P.V. 1984. The Evolution of Polynesian Chiefdoms. Cambridge: Cambridge University
Press.
Kirch, P.V. 2000. On the Road of the Winds: An Archaeological History of the Pacific Islands before
European Contact. Berkeley/Los Angeles/London: University of California Press.
Lee, G. 1992. The Rock Art of Easter Island: Symbol of Power, Prayers to the Gods, Monumenta
Archaeologia 17. Los Angeles: The Institute of Archaeology, University of California.
Lee, G. 2001. Te Moana Nui: Exploring Lost Isles of the South Pacific. Los Osos CA: Easter
Island Foundation/Bearsville Press.
Louwagie, G. and R. Langohr. 2002. Testing Land Evaluation Methods for Crop Growth
on two Soils of the La Prouse Area. Rapa Nui Journal, 16(1): 238.
319
Dynamics of Soil, Landscape and Culture on Easter Island
MacIntyre, F. 1999. Is Humanity Suicidal? Are there Clues from Rapa Nui? Rapa Nui
Journal, 13(2): 3541.
MacIntyre, F. 2001. Enso, Climate Variability, and the Rapanui, Part II: Oceanography
and Rapa Nui. Rapa Nui Journal, 15(2): 8394.
Martinsson-Wallin, H. 2002. Sea, Land, and Sky as Structuring Principles in Easter Island
Prehistory. Rapa Nui Journal, 16(2): 746.
Martinsson-Wallin. 2004. Archaeological Excavation at Vinapu (Rapa Nui). Rapa Nui
Journal, 18(1): 79.
McCall, G. 1993. Little Ice Age: Some Speculations for Rapanui. Rapa Nui Journal 7:
6570.
McCall, G. 1994. Rapanui, Tradition and Survival on Easter Island. Honolulu: University
of Hawaii Press
Mtraux, A. 1940, reprinted 1971. Ethnology of Easter Island. Honolulu: Bishop Museum
Press.
Mtraux, A. 1957. Die Osterinsel. Stuttgart: Verlag Kohlhammer.
Mieth, A., H.-R. Bork and I. Feeser. 2002. Prehistoric and Recent Land Use Effects on
Poike Peninsula, Easter Island (Rapa Nui). Rapa Nui Journal, 16(2): 8995.
Mieth, A. and H.-R. Bork. 2003a. Diminution and Degradation of Environmental Re-
sources by Prehistoric Land Use on Poike Peninsula, Easter Island (Rapa Nui). Rapa
Nui Journal, 17(1): 3441.
Mieth, A. and H.-R. Bork. 2003b. Land Degradation on Easter Island, Southern Pacific.
In Y. Li, J. Poesen and J. Zhang (eds), Gully Erosion under Global Change (Chengdu:
Sichuan Publishing House of Science and Technology), pp.18599.
Mieth, A., H.-R. Bork, W. Markgraf, I. Feeser and K. Dierssen. 2003a. Bodenerosion ein
Schlssel zum Verstndnis der Kulturgeschichte der Osterinsel. Petermanns Geographische
Mitteilungen, 147: 3037.
Mieth, A. and H.-R. Bork. 2005. History, Origin and Extent of Soil Erosion on Easter
Island (Rapa Nui). Catena, 63: 24460.
Mueller-Dombois, D. and F.R. Fosberg. 1998. Vegetation of the Tropical Pacific Islands.
Heidelberg/New York/Berlin: Springer Verlag.
Nunn, P. 2000. Environmental Catastrophe in the Pacific Islands around A.D. 1300.
Geoarchaeology 15(7): 71540.
Orliac, C. 2000. The Woody Vegetation of Easter Island between the Early 14th and
the Mid-17th Centuries AD. In C.M. Stevenson and W.S. Ayres (eds), Easter Island
Archaeology: Research on Early Rapa Nui Culture (Los Osos CA: Easter Island Founda-
tion), pp.21120.
Orliac, C. 2003. Ligneaux et Palmiers de lle de Pques du XI me au XVII me sicle de
notre re. In C. Orliac (ed.), Archologie en Ocanie insulaire peuplement, socits et
paysages (Paris: ditions Artcom), pp.18499.
Orliac, C. and M. Orliac. 1995. Easter Island, Mystery of the Stone Giants. New York: Harry
N. Abrams.
320
Andreas Mieth and Hans-Rudolf Bork
Orliac, C. and M. Orliac. 1998. The Disappearance of Easter Islands Forest: Over-ex-
ploitation or Climatic Catastrophe? In C.M. Stevenson, G. Lee and F.J. Morin (eds),
Easter Island in Pacific Context, South Seas Symposium, Proceedings of the Fourth
International Conference on Easter Island and East Polynesia (Los Osos CA: Easter
Island Foundation), pp.12934.
Orliac, C. and M. Orliac. 2001. Composition et volution de la Flore de LIle de Pques du
14 au 19s. Rapport sur les Travaux de la Mission Archologique Octobre Decembre
2000, Ministre des Affaires trangres, France, Consejo de Monumentos, Corporacin
Nacional Forestal, Chile Museum dHistoire Naturelle, Paris (unpublished).
Palmer, J.L. 1870. A Visit to Easter Island, or Rapa Nui, in 1868. The Journal of the Royal
Geographic Society, 40: 16580.
Ramrez, J.M. 2001. Cultural Resource Management of Easter Island: Utopia and Reality.
In C.M. Stevenson, G. Lee and F.J. Morin (eds), Pacific 2000, Proceedings of the Fifth
International Conference on Easter Island and the Pacific (Los Osos CA: Easter Island
Foundation), pp.38390.
Reichardt A. and I. Reichardt 2000. Osterinsel Ein Reisefhrer. Heidelberg: Max Kasparek
Verlag .
Rohdenburg, H. 1977. Beispiele fr Holozne Flchenbildung in Nord- und Westafrika.
Catena, 4(1/2): 65109.
Routledge, K. 1919, reprinted 1998. The Mystery of Easter Island. Kempton, Illinios: Ad-
ventures Unlimited Press.
Skottsberg, C. 1953. The Natural History of Juan Fernandez and Easter Island, Vol. 2, Botany,
19201953. Uppsala.
Skottsberg, C. 1956. The Natural History of Juan Fernandez and Easter Island, Vol. 1, Geog-
raphy, Geology, Origin of Island Life, 1920 1956. Uppsala.
Steadman, D.W., P. Vargas Casanova and C. Cristino Ferrando. 1994. Stratigraphy, Chro-
nology, and Cultural Context of an Early Faunal Assemblage from Easter Island. Asian
Perspectives, 33: 7996.
Stevenson, C.M. 1997. Archaeological Investigations on Easter Island, Maunga Tari: An Upland
Agricultural Complex. Los Osos, CA: The Easter Island Foundation.
Stevenson, C.M. and C. Cristino. 1986. The Residential Settlement History of Easter Islands
Southern Coastal Plain. Journal of New World Archaeology, 7: 2938.
Stevenson, C.M. and S. Haoa. 1998. Prehistoric Gardening Systems and Agricultural In-
tensification in the La Prouse Area of Easter Island. In C.M. Stevenson, G. Lee and
F.J. Morin (eds), Easter Island in Pacific Context, South Seas Symposium, Proceedings
of the Fourth International Conference on Easter Island and East Polynesia (Los Osos
CA: Easter Island Foundation), pp.20513.
Stevenson, C.M. and S. Haoa 1999. Diminished Agricultural Productivity and the Col-
lapse of Ranked Society on Easter Island. In P. Wallin (ed.), Lectures at the No Barriers
Seminar 1999 (Oslo: The Kon-Tiki Museum, Institute for Pacific Archaeology and
Cultural History), pp.412.
Stevenson, C.M., J. Wozniak and S. Haoa. 1999. Prehistoric Agricultural Production on
Rapa Nui. Antiquity, 73: 80112.
321
Dynamics of Soil, Landscape and Culture on Easter Island
Stevenson, C.M., Th. Ladefoged and S. Haoa. 2002. Productive Strategies in an Uncer-
tain Environment: Prehistoric Agriculture on Easter Island. Rapa Nui Journal, 16(1):
1722.
Sudzuki, F. 1979. Vegetacin Arbrea y Erosin en la Isla de Pascua. Boletn Tcnico 45
(Santiago: Facultad de Agronomia, Universidad de Chile): 38.
van Tilburg, J.A. 1994. Easter Island: Archaeology, Ecology and Culture. London: British
Museum Press.
Vogler, E. 1998. Die Europer entdecken die Osterinsel. In M. Esen-Baur (ed.), 1500 Jahre
Kultur der Osterinsel (Mainz: Verlag Philipp von Zabern), pp.5381.
Wallin, P., C.M. Stevenson and T. Ladefoged. 2005. Sweet Potato Production on Rapa
Nui. In C. Ballard, P. Brown, R.M. Bourke and T. Harwood (eds), The Sweet Potato in
Oceania: A Reappraisal, Ethnology Monographs 19 (University of Pittsburgh), Oceania
Monograph 56 (Sydney University), pp. 858.
Wozniak, J.A. 1998. Settlement Patterns and Subsistence on the North-west Coast of Rapa
Nui. In C.M. Stevenson, G. Lee and F.J. Morin (eds), Easter Island in Pacific Context,
South Seas Symposium, Proceedings of the Fourth International Conference on Easter
Island and East Polynesia (Los Osos CA: Easter Island Foundation), pp.18592.
Wozniak, J.A. 1999. Prehistoric Horticultural Practices on Easter Island: Lithic Mulched
Gardens and Field systems. Rapa Nui Journal, 13: 959.
Wozniak, J.A. 2001. Landscapes of Food Production on Easter Island: Successful Subsistence
Strategies. In C.M. Stevenson, G. Lee and F.J. Morin (eds), Pacific 2000, Proceedings
of the Fifth International Conference on Easter Island and the Pacific (Los Osos CA:
Easter Island Foundation), pp.91102.
Yen, D.E. 1974. The Sweet Potato and Oceania: An Essay in Ethnobotany, Bernice P. Bishop
Museum Bulletin 236. Honolulu: Bishop Museum Press.
Zizka, G. 1989. Naturgeschichte der Osterinsel. In M. Esen-Baur (ed.), 1500 Jahre Kultur
der Osterinsel (Mainz: Verlag Philipp von Zabern), pp.2138.
Zizka, G. 1991. Flowering Plants of Easter Island, Palmarum Hortus Frankofurtensis 3.
Frankfurt am Main: Palmengarten.
322
Frank Uekoetter
11
Frank Uekoetter
In the beginning, there was Liebig. That, at least, is the impression that one obtains
from agricultural publications in the twentieth century. When German farming
experts were discussing issues of plant nutrition and fertiliser use, and sought to
squeeze in some historical information in order to give depth to the narrative, it was
usually Liebigs name that shone the brightest. Of course, a good dose of patriotism
was sometimes fuelling references to the German professor, and certainly admiration
for Liebigs impressive career, from mediocre pupil to one of the most influential
scientists of the nineteenth century (cf. Brock 1997). But most of all, a reference
to Liebig seemed to be the perfect way to bestow more credibility upon ones own
argument. While Liebig fought bitter controversies with other scientists of his time,
retrospective references are almost exclusively of the complimentary kind.
However, the enduring appeal of Liebig to agricultural experts stands in stark
contrast to the disagreements over what he stood for. Many experts cherish him
as the founder of modern plant nutrition, as a textbook of 1951 noted (Schober
1951, p. 3). [Liebigs] mineral theory destroyed the old humus theory, a publication
of 1921 declared (Brandt 1921, p. 7). However, a third author invoked Liebig to
underscore his argument for the complete collection of human faeces in the cities
and their use as organic fertiliser (Deuen 1922, p. 826n). In the late 1920s, Max
Reiser warned of the neglect of agriculture and destruction of soil fertility with a
Liebig quote (Reiser 1929, p. 42). Johannes Grbing, who is today best known as a
pioneer of organic farming methods, used a quotation of Liebig admonishing us to
do our work with reverence to the Creation (Grbing 1947, p. 205). And in 2003,
an article on farm technology and sustainability referred to Liebig in arguing for
an economical use of minerals (Seufert 2003, p. 267). Obviously, Liebig, certainly
323
Know Your Soil
a controversial figure in his own lifetime, has somehow turned into everybodys
darling something of a free-floating icon, hovering over the agricultural science
profession, offering support to almost every faction in the field.
One could count these divergent readings of Liebig as mere curiosities if
they were not so typical of the profession in general. Ever since Liebig took on
Albrecht Thaers humus theory, there have been multiple scientific opinions in the
field (cf. Schling-Brodersen 1989 and Fussell 1971). Chemical, biological, bacte-
riological, geological and other approaches offered different perspectives on the soil,
and the clashes between these approaches were as intense as they were inevitable.
Interestingly, the competition between these approaches persists to the present day,
and with soil fertility being the result of both chemical and biological processes,
it seems unlikely that one approach will rise as unchallenged in the foreseeable
future. However, the ambivalence of scientific knowledge did not prevent the
chemical approach from emerging as the hegemonic one: from this point of view,
soil fertility and plant nutrition were synonymous with putting the right minerals
into the soil in the right quantities at the right time. The goal of agriculture is
[...] the proper and purposeful transformation of nutrients into plant substance
with the help of the soil as a mediator, a typical expression of this perspective ran
(Jahn-Deesbach 1965, p. 35). To be sure, the chemical approach never managed
totally to disprove or discredit the other approaches, but it did manage to push
these alternative approaches to the margins, transforming them into a source of
irritation for the farmer, rather than an inspiration for his daily work. Some of the
problems of modern industrial agriculture, like the contamination of groundwater
and the eutrophication of the landscape, are direct consequences of the hegemony
of the chemical approach.
Justus von Liebig offered a stinging critique of agricultural experimental
stations in his lifetime, claiming they were unable to conduct scientific research (cf.
Schling-Brodersen 1989, p. 235n). However, these experimental stations would
soon turn into key proponents of chemical conceptions of soil fertility, and as
early as 1913, three out of four experimental stations were headed by a chemist,
providing a head-start for agricultural chemistry that it would never really lose
(Reichrath 1991, p. 118). Nevertheless, it is important to realise that there were
always rivals to its hegemony. The early dominance of agricultural chemistry easily
leads to the impression that its hegemony was a natural phenomenon, and that
is how it comes across in the current literature: as an incontestable scientific em-
pire that rose in the nineteenth century and never showed any signs of weakness.
However, it becomes clear upon closer investigation that the path of agricultural
chemistry was far more complicated and twisted than that. Agricultural chemistry
was involved in constant battles; and these battles reached their peak during the
interwar years. The two decades between the world wars were a crucial time period
324
Frank Uekoetter
Of course, the same can be true for scientific knowledge, and it is just
as important to inquire more deeply into the rules and conducts of the relevant
branch of science: Who supplied the farmers with knowledge? Who defined the
good practice of farming and who did not? How did information flow from the
researchers to the individual farmers and what were the chances for feedback?
How have farmers a group with a legendary reputation for conservatism come
to accept advice on new farming methods so readily? Focusing on Liebig and his
opponents, researchers have frequently overlooked the complex set of agricultural
advisors who sought to bring scientific knowledge to the farming community.
Standing between the scientific profession and the farmers, the community of
agricultural advisors was where tensions and contradictions within the body of
agricultural knowledge would come to the forefront. Therefore, the agricultural
advisors deserve a prominent place in the history of agricultural knowledge in the
twentieth century, making the system of agricultural knowledge production and
diffusion even more complex.
All this is especially important given the background of recent arguments
that modern society has come to be dominated by knowledge-dependent processes
to such an extent that one could speak of a knowledge society.1 For all its merits,
it is important to realise that the theory of the knowledge society includes an
emphatic understanding of scientific research; characteristically, Nico Stehr argued
in his seminal work on the knowledge society that scientific research is increasingly
becoming the sole source of additional knowledge (Stehr 1994, p. 201). This focus
on the increasing influence of scientific knowledge easily leads to a point of view
that perceives every non-scientific kind of knowledge as a simple leftover from
previous times that is just waiting to be enlightened by scientific knowledge (cf.
Landwehr 2002, p. 17). However, non-scientific forms of knowledge persist even
in todays farming practice, and may in fact fulfil a number of important functions.
The rise of the agrarian knowledge society is not simply part of a Weberian process
of rationalisation (cf. Stehr 1994, p. 16).
1
Cf. the definition of Willke 1998, p. 162.
326
Frank Uekoetter
million tons (Mommsen 2002, p. 93). Therefore, the post-war years saw feverish
efforts to restore soil fertility and to remedy the effects of the wartime Raubwirtschaft
(exploitative economy) (Langenbeck 1921, p.8). However, the supply of fertilisers
had changed significantly since the 1910s. The invention of the Haber-Bosch proc-
ess made atmospheric nitrogen available for fertiliser production, thus opening a
practically limitless reservoir of the most critical plant nutrient (cf. Szllsi-Janze
1998). At the same time, manure had become scarce as a result of wartime cutbacks
on livestock and inadequate feed, which put the traditional role of organic wastes in
jeopardy (cf. Kuhn 1920, p. 347). As a result, mineral fertiliser quickly became the
method of choice: the almost unanimous hope of the scientific establishment was
that with massive doses of mineral fertilisers, pre-war yields would return almost
immediately. The exhaustion of our soils can be remedied only through an intensive,
rational use of mineral fertiliser, a brochure of 1921 declared (Smalakies 1921, p.
7). Increasing the use of mineral fertilisers became the rallying cry of practically
all advisors, and their efforts had the backing of both the scientific authorities and
the bureaucracy. In fact, a memorandum of the Prussian ministry of agriculture
expressed the urgency of the matter in a memorandum of 1920 as follows: The
necessary amounts of mineral fertiliser are available, and can be purchased and put
into the soil. If that does not happen, people will starve.2
However, the promise of a quick return to pre-war conditions turned out
to be deceptive. Per-acre yields stayed below pre-war figures for much longer than
expected, and the massive propaganda (to use a contemporary term that was not
yet tainted) in favour of mineral fertiliser came to appear dubious to many farm-
ers. Mineral fertiliser was the most significant investment for many farmers, and
with farmers struggling to keep out of debt during the 1920s, this issue was by no
means unimportant (Schlange-Schningen 1931, p. 23). In addition, the massive
use of mineral fertiliser led to acidification of many soils, which in turn inhibited
the growth of certain plants (Schellenberger 1924, p. 17). The acid condition of the
soil constitutes a threat that has attracted too little attention so far, a publication
of 1923 warned (Niggl 1923, p. 5). To be sure, farmers could use lime to restore a
neutral condition, but this implied additional costs, took another two or three years
with reduced yields, and provided the farmers with the irritating experiences that
it took a chemical substance to remedy a problem caused by a chemical substance
(Wagner 1929, p. 4). In fact, the sheer prospect of ruining the most essential basis
of agricultural production must have been disturbing, if not traumatic, for many
farmers. One expert later noted that the most important consequence of this episode
was the psychological effect: Agricultural chemistry lost credit in the eyes of many
2
Bundesarchiv R 3602 No. 606, Denkschrift des preuischen Landwirtschaftsministers zur
Frage der Volksernhrung, November 1, 1920, p. 1.
327
Know Your Soil
3
Bundesarchiv R 3602 No. 67, Niederschrift der 80. Sitzung ber allgemeine Dngerangele-
genheiten im Preuischen Landwirtschaftsministerium am 13. Mai 1931, p. 14; Niederschrift
der 81. Sitzung ber allgemeine Dngerangelegenheiten im Preuischen Landwirtschaftsmin-
isterium vom 13. Mai 1932, p. 23.
4
Bundesarchiv R 3602 No. 608, the director of the Biologische Reichsanstalt to the Re-
ichsgesundheitsamt, January 10, 1935.
5
Similarly Schnelle and Nolte 1933, p. 223.
329
Know Your Soil
from this straw fire, one expert wrote (Flieg 1933, p. 715). But if that was so, why
did these experts not simply relax and wait until the anthroposophists had worked
their way into bankruptcy? Why did the agricultural science establishment react
with so much vigour, and even go to the extreme of trying to solve the dispute with
the help of the Nazi regime? To a certain extent, these attacks were a response to the
way that anthroposophists were marketing their products. Organic farmers often
focused on the public at large, rather than other farmers, in their publicity efforts,
suggesting a link between the use of mineral fertiliser and cancer (Scharrer 1934,
p. 245). The fear of cancer turned out to be a powerful marketing tool, and many
consumers sought organic farming products as a result. Numerous farmers were
selling their products as grown without mineral fertiliser in order to achieve better
revenues, even though they had followed the conventional methods of fertilising;
certification of organic farming products, and even clear definitions of the dos and
donts of organic farming, were still several decades away, thus the door was open
for a lot of fraud under the organic farming label (Flieg 1933, p. 715). Of course,
all this was bound to evoke no sympathy for organic farming in the agricultural
science establishment, and yet it would be misleading to depict the massive attacks
on organic farming as simply a response to a challenge from outside. In its critique
of organic farming, the scientific establishment was also pursuing a number of
strategic goals. In the light of their interests, and against the background of the
crisis of confidence of the 1920s, one could even argue that the clash over organic
farming was a golden opportunity to hammer home a few important points.
On the most general level, the critics of organic farming sought to prevent
the formation of a competing network of expertise. The campaign did not fully suc-
ceed in this regard, as the continued existence of organic farming up to the present
shows, but it did succeed in confining organic farming methods to a small fringe.
Even more, the critics succeeded in preventing any meaningful exchange between
conventional and organic farming: every practitioner knew that he could turn to
either system of expertise, but not to both at the same time. Even on a scientific level,
contacts were rare, and were met with great scepticism; up to the 1980s, seeking a
dialogue with organic farming experts put a scientists professional reputation at risk
(Gerber et al 1996, p. 596). As a second welcome side-effect, the critique of organic
farming led to a general scepticism towards biological concepts of soil fertility; in
effect, the controversy canonised a chemical understanding of soil fertility. The true
importance of this impact becomes clear when one looks at the general character
of the field. The agricultural chemistry establishment was a chaotic conglomera-
tion of research institutions from completely different backgrounds: university
departments and institutes sponsored by the farmers (Landwirtschaftskammern)
coexisted happily with institutes run by the producers of mineral fertilisers. The
importance of the latter becomes apparent when one considers the major producers
of mineral fertiliser in Germany: potash fertilisers came from state-owned mines,
330
Frank Uekoetter
heavy industry sold phosphate from the production of steel in Thomas converters,
and IG Farben supplied nitrogen fertilisers the three most powerful forces of
the German economy, united by their interest in increasing fertiliser sales! Against
this background, the lack of a truly independent branch of agricultural chemistry
is by no means surprising.
However, the complex mixture of industrial, agricultural, and scientific
concerns also implied a considerable weakness of this research network: rules of
good conduct were notoriously difficult to establish. There was no supreme authority
that could define certain rules, and neither was there a place with Peter Galison
(1995), one might also speak of trading zones where the profession could meet
and develop rules of this kind. Therefore, the conflict with organic farming provided
a perfect opportunity to establish a few important rules: dont get in touch with
organic farming experts, dont seek an understanding of soil fertility that moves
beyond chemistry, and dont bother too much about organic fertilisers. Of course,
every expert in the field was aware of the fact that the vast majority of farmers
used manure on a regular basis but at the same time, the general tendency in the
field was to marginalise the study of manure as a fertiliser, and to focus on mineral
fertiliser instead. Over time, this turned into a highly consequential bias: mineral
fertiliser came to be regarded as the true fertiliser that commanded the lions share
of attention while organic wastes were relegated to the status of a marginal topic.
As a result, manure did not attract major attention from the agricultural science
establishment until long after the Second World War and then, unexpectedly,
manure became an issue not in the form of the asset that it actually was, but,
strangely enough, in the form of an environmental problem.
3. Trust in Numbers:
The Beginnings of a Hazardous Game
While agricultural chemistry was competing with other schools of soil fertility, it
was simultaneously struggling to resolve a second problem that was instrumental
in its rise to dominance: the issue of dosage. The conventional literature provided
ample information about the characteristics of the individual nutrients, the types
of fertilisers, and their proper handling. However, as farmers became more reliant
on mineral fertiliser during the interwar years, they were no longer content with
general information. Faced with a crisis of confidence and the significant economic
risk of high fertiliser expenditures, they were asking for more precise directions:
what amount of what type of fertiliser should they use for their crops? From the
point of view of agricultural scientists, this posed a tricky dilemma: they could not
ignore this demand without risking their own jurisdiction over fertiliser use but
at the same time, they had no scientifically proper method to meet it.
331
Know Your Soil
6
E.g. Arbeitsgemeinschaft der deutschen Stickstoff-Industrie fr das landwirtschaftliche
Beratungswesen 1937, p. 57.
332
Frank Uekoetter
However, this hesitancy gradually disappeared, and that paved the way for
todays system of soil analysis: a laboratory of the state-funded agricultural research
institute (Landwirtschaftliche Untersuchungs- und Forschungsanstalt) analyses the soil
samples that farmers need to submit at least every six years, and for those farmers
who submit an overview of future crops to the institute beforehand, a computer
program with the charming name DungPro produces a print-out with detailed
information on the necessary nutrients (cf. Jacobs and Remmersmann 2003). This
routine procedure is all the more remarkable since the initial uncertainties of the
method have by no means disappeared. In fact, a recent handbook on plant nutrition
and fertilising praised soil sampling for its speed and low costs but also mentioned
the findings low precision as a disadvantage (Schilling 2000, p. 294). The causes
of error are numerous: acres are usually not uniform in their mineral content, a fact
that soil samples usually ignore in favour of an arithmetic average; the method does
not account for the concentrations in different depths; the solvent needs to extract
within a matter of hours what plants would extract over the span of an entire year;
and the method inevitably needs to ignore the impact of future weather conditions
(Schilling 2000, pp.294, 299). All in all, the uncertainties are staggering.
However, scientific precision is not necessarily the most important aspect
for the selection of a certain method. In the everyday consultation business, time
soon came to dominate the choice of methods. The experience of Eilhard Alfred
Mitscherlich with soil investigations in Eastern Prussia provides a case in point.
Faced with the challenge of writing some 2,000 reports within a brief span of time,
he came to use a simple form that included exact figures and postponed detailed,
non-quantitative comments to personal meetings during the winter (Mitscherlich
1931, pp. 1114). The farmers persistent demand for exact figures met with a
growing readiness of scientists to supply this kind of information, and that cre-
ated a dynamic of its own that gradually pushed all scientific doubt to the mar-
gins. Interestingly, Mitscherlich took a much more sceptical view of this method
two decades later: It is certain that one will never be able to tell the farmer how
much of a certain fertiliser he should give on the basis of chemical soil analyses
(Mitscherlich 1952, p. 7). However, objections of this kind did little to stop the
rise of this method.
Today, the advisory literature generally ignores all the uncertainties that
chemical soil analyses include. Soil analyses are the indispensable tool for optimis-
ing fertiliser use, a recent article in a popular farming journal declared (Jacobs and
Remmersmann 2003, p. 32). The environmental contingencies of this approach
become apparent when one realises that even if the figures were precise, the approach
would inevitably ignore a whole host of factors. Condition and size of the humus
layer, bacterial activity, density of the soil these and other parameters naturally
have an influence on soil fertility, but they do not enter into the calculations of
DungPro, which simply sees the soil as a temporary depot of minerals. In fact,
333
Know Your Soil
even a severe problem like erosion will remain undetected within this system of
knowledge a worrisome situation in a country where erosion is usually a long-term
process without conspicuous events of the dust bowl kind. Erosion is not simply an
environmental problem in German agriculture. It is also something that happened
to the body of agricultural knowledge.
7
For the recent interest in trust as a historical phenomenon, see Frevert 2003.
8
For a telling expression of this line of reasoning, see Planck 1985.
334
Frank Uekoetter
However, trust is only one part of the explanation. The other part is distrac-
tion the preoccupation of todays farmers with other issues. It is by no means
coincidental that the increasing dominance of scientific advisors went parallel with
the mechanisation of agricultural production: the former was possible only because
from the farmers standpoint, it was embedded in the latter. As in the field of fertiliser
use, the early relationship between the farmers and their machines went through
a period of estrangement. In the 1920s, one author reported that many farmers
still saw technology as a necessary evil (Endres 1926, p. 256). Similarly, Hans
Schlange-Schningen (1927, p. 42) mocked the tractor-fancy as a showcase of
the exaggerations of modern farming. But soon machines became a routine part of
agricultural work, and ultimately a source of pride. A new type of farmer emerged:
the technologist with a penchant for up-to-date machinery and do-it-yourself repairs.
In the 1990s, agricultural machinery dealers were selling (as opposed to installing)
about 50 percent of the total number of spare parts because the farmers wanted
to install these parts themselves (Thiede 1992, p. 65n). Of course, saving money
was part of the motivation, but so was a fancy for technology. Interestingly, the
advisory literature no longer pushes the use of the latest machinery but rather calls
for caution in this respect. Those who are buying the latest technology for reasons
of prestige are usually producing with excessive costs, a recent handbook on farm
economy noted (Mohn 1995, p. 139). However, the persistent rise of horsepower
figures of farm tractors since the Second World War provides a telling documenta-
tion of the farmers priorities.
The links between the growing prominence of scientific experts and the
mechanisation of farm production were functional as well as ideological. On a
functional level, machines required a lot of attention and time for maintenance,
thus diminishing the farmers chance to monitor other parts of their enterprises as
diligently as before. For example, while farmers traditionally learned a lot about
their soils by walking behind the plough, farmers using tractors for tillage devoted
much of their attention to the proper operation of the engine. Thus, an expert who
offered detailed information on proper fertiliser use could appear not as an intruder
into a core area of farming knowledge but as a welcome aide who brought clarity
into an area which one could not deal with thoroughly for lack of time.
However, the impact of mechanisation on the farmers identity was prob-
ably even more important. If a powerful tractor and up-to-date equipment was the
farmers new fetish, it was consistent that they became uninterested in the issue
of fertiliser; after all, chemicals and odorous wastes are about the least prestigious
farming tools that one can imagine. As a result, farmers readily ceded authority on
this dull subject to a trustworthy expert; for them, the key issue was not autonomy
of decision, but prestige. Naturally, it would take a major incentive, or incident,
for these farmers to redirect their attention to the soil.
335
Know Your Soil
5. Conclusions
One of the surprising features of todays system of agricultural production is its
remarkable resilience to criticism. Reading Hermann Priebes Subsidised Nonsense
almost twenty years after its publication, one is struck by the enduring validity
of much of what he wrote. In fact, it seems that there are few fields where the
environmental movement has made as little headway as in the field of agriculture.
Obviously, chastising farmers for their environmental toll, or urging them to convert
to organic farming, has been less effective in the field of agriculture than analogous
criticism of industrial corporations. In fact, it seems that this kind of criticism raised
tempers far more than it raised awareness of environmental problems.
In light of the previous discussion, an important reason for this dismal situation
is now clear: the environmentalists critique ignores the fact that the environmental
problems of industrial agriculture are intrinsically connected to a knowledge system
that farmers cannot easily abandon. Having surrendered jurisdiction over soil fertility
and plant nutrition to outside experts over the course of the twentieth century, the
key problem is not a lack of good intentions on the side of the farmers but a lack
of knowledge systems. Under an advisory regime that ignores any non-numerical
input, the farmer has little choice but to accept the DungPro print-out and follow
its directions and the preoccupation with the latest machinery keeps most farm-
ers from thinking about their fateful situation. And if a farmer chooses to distrust
the experts recommendations, that leaves him with the even more dismal task of
deciding autonomously. In their book on the history and future of the soil, John
Seymour and Herbert Girardet (1985) repeatedly note that farmers are heeding
scientific advice while at the same time being uneasy about it. This may look like
irrational behaviour at first glance until one considers the available alternatives.
If farmers are heeding their advisors recommendations, this is not due to the aura
of scientific expertise, or to ignorance about the narrow approach that the experts
are taking. In many cases, farmers simply trust for lack of a choice.
Interestingly, even the reforms of recent years have done little to change this
situation. Renate Knast, the German minister of agriculture who came into office
in early 2001 in the wake of the BSE scandal, devoted much energy to the promo-
tion of organic farming, hoping for a 20 per cent market share in the foreseeable
future. Aside from the still open question whether there is really a sizable market
for organic farming products prices for organic farming products have dropped
significantly in recent years, indicating an overproduction even with the present
market share18 this approach has tended to decrease pressure on conventional
18
Cf. Budde 2003. In 2000, 3.2 per cent of Germanys farmland was under cultivation
pursuant to the rules of organic farming. (Schmidt and Jasper 2001, p. 111.)
336
Frank Uekoetter
farmers to improve their environmental record, and in fact tends to depict the
system of conventional farming as a dated model that farmers can only abandon,
but not change in an environmentally responsible way. To be sure, some reformers
have developed a complex network of financial incentives and legal prohibitions
that seek to spread certain practices and ban others. However, there seems to be
a strange reluctance to put the issue of agricultural knowledge centre stage: who
is defining the good practice of farming, according to which criteria, with what
overall goal? Not least, a reform of the agricultural knowledge system would have
to consider ways to incorporate the farmers experiences in a productive way. At
present, it seems likely that the new system of agricultural knowledge will margin-
alise the experiences of the farmers in the same way that the old one did. Farmers
will be no less helpless in dealing with the new environmental canon than they are
now under the regime of DungPro.
In his monograph on the knowledge society, Nico Stehr notes that a
knowledge society differs from previous types of society in that its structure is to
a greater extent the result of social action. It is a society where secondary nature
is vastly more important than primary nature, Stehr notes (1994, p. 218). An
environmental historian will certainly read such a statement with a sense of alarm:
what happens if the rules of the knowledge society are at odds with the dynam-
ics of nature? From such a point of view, the agrarian knowledge society appears
like a runaway steam train a system of knowledge with a vast impact on nature
and little room for feedback. The prospect of this situation should be all the more
reason to put the current system of agricultural knowledge under scrutiny and
to start an open discussion on the knowledge base of modern agriculture. The
history of agricultural knowledge in the twentieth century shows that this will
demand a lot from all parties involved, but also that such a discussion may easily
gain momentum once it has started. After all, agricultural knowledge has always
been pluralistic, even at times when agricultural chemistry seemed uncontested at
first glance. Promoting a plurality of opinions, and of approaches, is an important
contribution to the quest for an agriculture that is both productive and sustainable.
The call is for scientists to communicate with both farmers and their perceived rivals
in other scientific branches and to combine these divergent contributions into
the vision of a new agriculture. In other words, the agricultural knowledge society
will need to turn into an agricultural learning society. And researchers should not
be discouraged by the fact that there is probably no Liebig quotation to legitimate
this endeavour.
337
Know Your Soil
References
Arbeitsgemeinschaft der deutschen Stickstoff-Industrie fr das landwirtschaftliche Beratungs
wesen. 1937. Dngungsratschlge fr den Bauernhof, 2nd edition. Berlin: Arbeitsgemein-
schaft der deutschen Stickstoff-Industrie fr das landwirtschaftliche Beratungswesen.
Brandt, O. 1921. Gemeinverstndliche Einfhrung in die Stickstoffrage. Hannover: Engelhard
& Co.
Brock, William H. 1997. Justus von Liebig: The Chemical Gatekeeper. Cambridge/New York:
Cambridge University Press.
Budde, Franz-Josef. 2003. Biobauern triffts gleich mehrfach. Landwirtschaftliches Wochenblatt
Westfalen-Lippe, no. 38, 18 September 2003: 11.
Deuen, E. 1922. Kritisches zu unserer Dngernot, nebst einem Beitrag zum Phosphorgehalt
der Jauche. Chemiker-Zeitung, 46: 8268.
Endres, W. 1926. Maschinentechnische Hilfsmittel des Landwirts. In F. Christoph (ed.),
Die Technik in der Landwirtschaft (Berlin-Schmargendorf and Leipzig: Andermann),
pp. 25688.
Flieg, O. 1933. Hnde weg von unseren bewhrten Dngemethoden! Die Umschau. Il-
lustrierte Wochenschrift ber die Fortschritte in Wissenschaft und Technik, 37: 7135.
Frevert, Ute (ed.). 2003. Vertrauen. Historische Annherungen. Gttingen: Vandenhoeck
& Ruprecht.
Fussell, George E. 1971. Crop Nutrition: Science and Practice before Liebig. Lawrence:
Coronado Press.
Galison, Peter L. 1995. Context and Constraints. In Jed Z. Buchwald (ed.), Scientific Prac-
tice: Theories and Stories of Doing Physics (Chicago and London: University of Chicago
Press), pp. 1341.
Gerber, Alexander, Volker Hoffmann and Michael Kgler. 1996. Das Wissenssystem im
kologischen Landbau in Deutschland. Zur Entstehung und Weitergabe von Wissen
im Diffusionsproze. Berichte ber Landwirtschaft, 74: 591627.
Gerlach, Max. 1911. Langjhrige Feldversuche ber die Wirkung der Kalisalze auf Sandbden
(Landwirtschaftliche Lehrhefte no. 5). Berlin: Verlagsgesellschaft fr Ackerbau.
Glathe, H. 1935. Vorkommen und Ttigkeit von Mikroorganismen im Boden. In F. Lh-
nis et al (eds), Handbuch der landwirtschaftlichen Bakteriologie. Band II: Dnger- und
Bodenbakteriologie (2nd rev. edition, Berlin: Borntraeger), pp. 159629.
Grbing, Johannes. 1947. Die Grundlagen der Gare im praktischen Ackerbau, vol. 1. Han-
nover: Landbuch-Verlag.
Hartmann, Friedrich. 1930. Kalkfibel in Frage und Antwortstil. Coburg: Hofbuchhandlung
A. Seitz.
Hppner, Gustav. 1911. Die Kunstdngemittel und ihre Anwendung in der modernen Land-
wirtschaft (Lbes Landwirtschaftliche Bibliothek vol. 28/29). Leipzig: Reichenbachsche
Verlagsbuchhandlung.
Jacobs, Gnter and Theo Remmersmann. 2003. Erst untersuchen, dann dngen. Land-
wirtschaftliches Wochenblatt Westfalen-Lippe, no. 31, 31 July 2003: 324.
338
Frank Uekoetter
Planck, Ulrich. 1985. Die Landwirtschaft in der Industriegesellschaft und die Industri-
alisierung der Landwirtschaft. Zeitschrift fr Agrargeschichte und Agrarsoziologie, 33:
5677.
Priebe, Hermann. 1985. Die subventionierte Unvernunft. Landwirtschaft und Naturhaushalt.
Berlin: Siedler.
Reichrath, Susanne. 1991. Entstehung, Entwicklung und Stand der Agrarwissenschaften in
Deutschland und Frankreich. Frankfurt: Peter Lang.
Reiser, Max. 1929. Die Wiesen- und Weidewirtschaft im Allgu unter Bercksichtigung
der Gllewirtschaft. In Verein der Thomasmehlerzeuger (ed.), Unseren Freunden vom
Thomasmehl (Berlin: Verlag fr Bodenkultur), pp. 3242.
Remmersmann, Theo. 2004. Der Computer hilft bei der Dngeplanung. Landwirtschaftliches
Wochenblatt Westfalen-Lippe, no. 4 , 22 January 2004: 2830.
Scharrer, K. 1934. Die biologisch-dynamische Dngung im Lichte der Agrikulturchemie.
Chemiker-Zeitung, 58: 2457, 26770.
Schellenberger. 1924. Zeitgeme Vorschlge fr die Dngung der landwirtschaftlichen Kul-
turpflanzen (Arbeiten aus dem Gebiete der schsischen Landwirtschaft no. 7). Dresden:
Verlag des Landeskulturrats Sachsen.
Schilling, Gnther. 2000. Pflanzenernhrung und Dngung. Stuttgart: Ulmer.
Schlange-Schningen, Hans. 1927. Rationalwirtschaft und Nationalwirtschaft. Betrachtungen
eines praktischen Landwirtes. Berlin: Reinhold Khn.
Schlange-Schningen, Hans. 1931. Das Wirtschaftsjahr des praktischen Landwirts. Berlin:
Parey.
Schling-Brodersen, Ursula. 1989. Entwicklung und Institutionalisierung der Agrikulturchemie
im 19. Jahrhundert. Liebig und die landwirtschaftlichen Versuchsstationen (Braunschweiger
Verffentlichungen zur Geschichte der Pharmazie und der Naturwissenschaften 31).
Stuttgart: Deutscher Apothekerverlag.
Schmidt, Gtz, and Ulrich Jasper. 2001. Agrarwende oder die Zukunft unserer Ernhrung.
Mnchen: Beck.
Schneider, Johannes. 1924. Dngung. In Johannes Schneider (ed.), Praktisches Handbuch
der Landwirtschaft (Reutlingen: Enlin & Laiblins Verlagsbuchhandlung), pp. 5394.
Schnelle, G. and O. Nolte. 1933. Ueber die sogenannte Reformbewegung fr Umstellung
der Volkskost. Fortschritte der Landwirtschaft, 8: 2203.
Schober, Karl. 1951. Dngerlehre. Ein Handbuch fr praktische Landwirte und Studierende.
Wien and Heidelberg: Bohmann.
Seufert, Hermann. 2003. Nachhaltige Landtechnik im Sinne Liebigs. In Der Prsident der
Justus-Liebig-Universitt Gieen (ed.), Justus Liebig (18031873). vol. 1: Seine Zeit und
unsere Zeit. Chemie - Landwirtschaft Ernhrung (Gieen: Justus-Liebig-Universitt),
pp. 26776.
Seymour, John, and Herbert Girardet. 1985. Fern vom Garten Eden. Die Geschichte des Bodens.
Kultivierung - Zerstrung Rettung. Frankfurt: Wolfgang Krger Verlag.
Smalakies, Arthur. 1921. Die Bedeutung der Kunstdnger fr den Wiederaufbau der deutschen
Landwirtschaft. Berlin: Elsner.
340
Frank Uekoetter
Stehr, Nico. 1994. Arbeit, Eigentum und Wissen. Zur Theorie von Wissensgesellschaften.
Frankfurt: Suhrkamp.
Stremme, Hermann. 1926. Grundzge der praktischen Bodenkunde. Berlin: Borntraeger.
Szllsi-Janze, Margit. 1998. Fritz Haber 18681934. Eine Biographie. Mnchen: Beck.
Thiede, Gnther. 1992. Die Grne Chance. Landwirte zwischen Tradition und Fortschritt.
Frankfurt: DLG-Verlag.
Wagner, Paul. 1920. Anwendung knstlicher Dngemittel (Thaer-Bibliothek no. 100). 7th
edition, Berlin: Parey.
Wagner, Paul. 1929. Das Thomasmehl als Frhjahrsdnger. In Verein der Thomasmeh-
lerzeuger (ed.), Unseren Freunden vom Thomasmehl (Berlin: Verlag fr Bodenkultur),
pp. 36.
Wiessmann, H. 1926. Die Bodenaciditt und ihre Bedeutung fr das Pflanzenleben.
Zeitschrift fr angewandte Chemie, 39: 52530.
Willke, Helmut. 1998. Organisierte Wissensarbeit. Zeitschrift fr Soziologie, 27: 16177.
Archival Sources
Bundesarchiv R 3602 No. 67.
Bundesarchiv R 3602 No. 606.
Bundesarchiv R 3602 No. 608.
341
Index
Index
A agricultural treatises, 13
agriculture, 1345, 66, 247, 322336
Abeokuta, 149 as dominant concept in society, 177
Aboriginal people, 252 beginnings of, 15
absorption bank, 255 drained/ridged field, 108, 109
Abul Fazl, 30 failure of modern, mechanised, 65
acidification, 160, 221 flood recessional, 103, 105, 108, 111
adamellite, 264 Harappa, 17
adaptation, 185, 243, 249, 250, 251, 258 indigenous, 65
advisors, agricultural, 325. See alsofarm advisors industrial, 247. Seeindustrial agriculture
aeration change by ridges, 73 intensive, 56, 70
aerial photography, 94, 112 irrigated, 260
identification of gilgai, 99 lectures on, 328
aerobic microbial communities, 95 Maya, 69
afforestation, 27, 28, 44, 45 mechanisation of, 334
Afghanistan, 27, 28 opportunistic, 36
Africa, 4, 118176. See alsoCentral Africa, East raised field, 109, 110
Africa, West Africa revenue from, 24
southern, 135, 151, 155, 156, 160, 162 rise of African, 134
African soils, 118, 125 slash-and-burn, 25, 56
measurable properties, 121 spread of, across Europe, 9
agave, 62, 74 swidden, 53
Agdell rotation, 235 transitions to, 1
aggradation. Seesoil processes;See alsosediment wetland
accumulation inputs, 109
Agra, 30, 32 intensive, 111
agriculturalists, 134 agroecosystems, 41, 42
agricultural chemistry, 323, 327330, 333, 336 balanced, 41
agricultural development equilibrium, 45
errors of, 97 integrated, 72
agricultural engineers, 158 lost, 45
agricultural exhaustion, 65 agroforestry, 135, 306
agricultural experiments, 260 sustainable, 304
agricultural experiment stations, 155, 157, 323 transformation to horticulture, 308
agricultural literature, 217. See alsotextbooks Aguateca, 105
agricultural officers, 263 A-horizon, 244, 260
agricultural practice, 19, 46 Ah-horizon, 293, 298, 300
agricultural production, 139 Ah/Ap-horizon, 300
Agricultural Revolution, 238 Ah/B-horizon, 297
agricultural scientists, 46 Ahmedabad, 32
agricultural societies Aiel, 135
mentality, 179 Ain-i-Akbari, 30, 31, 32
worldview, 184
342
Index
compost, 28 D
Congo, Democratic Republic of, 152, 154
Congo River, 126 daga, 149
conservation, 46, 159 Dalmatia, 239
agencies, Easter Island, 289 dams, 57, 252, 254, 255, 258, 260
campaigns, 159 wooden or earthen, 95
ethic, 13 Danangombe, 149
management, 314 Dara Shikoh, 32
practices, 158 date palm, 135
Constantine VII Porphyrogenitus, 194 Davy, Sir Humphrey, 37
contamination, groundwater, 323 deforestation, 44, 57, 69, 77, 91, 101104,
contour 111, 123, 124, 157, 161, 162
banks, 158 Easter Island, 285, 305, 310
ridges, 158 Himalayan, 44
conucos, 75, 76 impact on runoff and water table, 111
Cook, James, 280 pre-Hispanic, Carribean, 76
cooling water, 33 degradation, 123
Copn, 65, 66, 70, 72, 103 Preclassic, 69
copper, 20, 62, 152, 154, 162, 163 Deiotarus, 192
corn, 30, 53 Delhi, 30
corn pollen, 53 demand, 7
corralling, 160 Demeter (Ceres), 179
cotton, 40, 56, 76, 157, 282 Denmark, 219
coupled systems, humanenvironment, 92 denudation of uplands, 218
Covent Garden, London, 238 deposition, 8
cradle of humanity, 133 deposits, outwash, 218
Cretaceous, 59 desertification, 122
crisis narratives, 122, 164 deshret, 134
cropping systems development. Seeagricultural development,
effect on erosion, 236 economic development
rotational, 136 de Landa, Diego, 61, 74
crop diseases, 226 Diana, 180
crop production, 135147, 159, 160 diatoms, 63
African, 121 Didymus of Alexandria, 194
crop rotation, 10, 11, 224, 232 digging stick, 56, 65, 293
experiments, 234 Dionysius of Utica, 192
in 18th century Europe, 217 Dionysos (Bacchus), 180
crusting, 125 Diophanes of Bithynia, 192
Cuba, 53, 54, 56 discourse, 186
Culavamsa, 34 diseases, 233. See alsofungal diseases
cultivated land, 17, 19, 25, 29, 31, 44, 45 foliar, of cereals, 224
cultivation, 7, 134 import to Easter Island, 280
capability for, 31 susceptibility of crops to, 233
limited by soil properties, 25 ditched fields, 92
cultural consequences of landscape and soil ditching, 56
change, 308 reasons for, 96
cultural systems diversion
relation to natural systems, 92 bank, 255
customs, agricultural, 3 ditches, 158
cyanobacteria, 73 weirs, 76
Doab, 32
347
Index
H prehistoric, 288
retreat of, 310
Haber-Bosch Process, 326 subsistence, 57
haematite, 154 house mounds, 105
Haiti, 76 Hrabanus Maurus, 197, 201
catastrophic level, 78 Hulbert Creek, Wisconsin, 110
Hamilton, Francis Buchanan, 29, 36, 37, 123 human-environment relationship, 251
Hampshire, 225 human-induced change, 64
Hanga Roa, 286 humans as soil forming factors, 134
Harappa, 16, 17, 18, 50 human agency, 44
collapse, 17 human health, 13
Harappan period, 15, 16, 17, 18, 19, 21, 22 human impact, 249
Hardman Syncline, 259264, 269 human response to change, 245
hardpans. See ironpans Humboldt current, 305
Harmattan winds, 125 humic horizon, 298
harvesting, 19 humour theory, 206
Haryana, 18, 38 humus, 7, 322, 323, 328, 332
Hatton Garden, London, 238 content of soils, 294
Hawaii, 274, 275 dating of, 287
hawaiite, 287 horizon, 298
hay, 238 hunter-gatherers, 15, 53, 133, 134, 217
Hazar Tuki River, 22 hunting-fishing-gathering economies, 54
herbicides, 165 Hutton, James, 35, 36
herbivores. Seeanimals hydroelectric power schemes, 255, 259
Heriger of Lobbes, 183 hydrologic change, 52
Hesiod, 179, 193, 194, 217 hydrology, 165
Heyerdal, Thor, 279 link to history of soils, 92
Hieron, 192 surface, 159
Himachal Pradesh, 39 hydrophytes, 91
Himalayas, 17, 31, 3841, 44, 45, 49 Hydro Agri, 333
Hinduism, 26, 32
Hindus, 24 I
Hispaniola, 53, 54, 56, 75, 76
historians, 46 Ibadan, 149
histosol, 59, 95, 103 Iberian peninsula, 194
Holocene, 8, 52, 53, 70, 133, 287, 300 Ibn Al Awwm, 194
arable agriculture, 221 Iburg, Benedictine monastery, 179
early, 219 ice age, 8, 9, 218
Late, 94 IG Farben, 330
Homer, 2 Illinois, 109
home economics, 183 inceptisol, 59, 95
Hoos Field plots, 233, 235 incision, 244, 252, 254, 257259, 267, 268
Hppner, Gustav, 331 India, 1550, 123
horticulture, 280, 281, 288, 298, 300, central, 14
303305, 310, 315 Mughal, 32
abandonment of, 302 northern, 14, 27, 28, 34, 36
Easter Island, 274, 293, 297 peninsular, 125
forest-related, 294 southern, 14, 19
increase in, 281 ancient, 43
intensification, 281, 308 Indiana, 109
labour-intensive, 312 indigenous flora and fauna, 283
indigenous knowledge, 60
352
Index
NSW Soil Conservation Service, 258 Ord River, 259, 260, 262
NSW Water Conservation and Irrigation catchment, 259, 268, 269, 270
Commission, 256 Dam, 261
Nushka Dar Fanni-Falahat, 28, 32 Ord River Regeneration Reserve, 262
nutrient Ord River Station, 260
availability, 40 organic carbon, 161
in root zone, 217 organic farming, 11, 239, 322, 328, 329, 330,
cycling, 138139, 143145, 164165 335
depletion, 57 in Europe, 217
imbalance, 229 organic matter, 7, 9, 10, 11, 36, 78, 95, 156,
leaching, 9 160, 165, 217, 220222, 228, 229,
loss, 122, 160, 230 236239, 247
long-term, 2 content of soil, 232
management, 10, 268 high, 221
mining, 122 low, 234
recycling, 41, 96 degradation, 222
relocation, 223 loss, 234
removal by forage, 223 reduction, 235
reserves, 219 sulphide catalysed oxidation of, 100
supply, 7, 96 organic wastes, 324
nutrients, 7 Organization of African Unity, 129
in top soil, 77 Orinoco River, 107
Nyasaland, 157, 158 Basin, 54
Orissa, 27, 34
O orogenic episodes, 218
Orongo, 279
oak, 220 Orstrom classification, 130
oats, 10, 216, 218, 224, 225, 231, 234 Ostionoid. SeeCeramic
yields of, 234 Otjimbingwe, 160
Oaxaca, 65 Ottoman empire, 32
observation, 127 Oudh, 34
obsidian outwash sand, 8, 219
tools, 301 Ovakwanyama, 163
weapons, 278 Ovambo plain, 163
Oceania, western, 275 overgrazing, 274, 302, 311
Odyssey, 2 overpopulation, 275
Oeconomica (Coler), 209 oxisol, 65
Ohio River, 109 oxygen
Oikonomikos, 193 isotopes, 91
oikos, 183, 184 isotopic analysis, 63
oil palm, 135 Oyster Harbour, 264, 266, 267, 268, 269
Old World diseases, 75
oligotrophic conditions, 9 P
olive trees, 32
Olmec, 54, 99 Pacific region, central, 275
Oneota people and pottery, 110 paddy rice ecosystems, 96
onions, 108 Pakistan, 15, 28
oral history, 119, 161 northern, 14, 26
Orange Free State, 156 western, 15
Orange River, 149, 163 palaeoecologists, 46
Orapa, 162 Palenque, 103
Ord Irrigation Scheme, 262 paleoecological proxies, 99
358
Index
paleosol, 68, 69, 91, 100, 101 peat, 59, 95, 100, 103, 237
datable, 62 pedogenesis, 129
deep, 101 pedological research, 244
weathered, 232 pedotransfer functions, 130
Yucatec Mayan, 68 Pelsaert, Francisco, 31, 33
palimpsest, 57 peneplanation, 124
Palladius, Aemilius Taurus, 186 peppers, 76
Palladius, Rutilius Taurus Aemilianus, 186, Periods of caribbean history, 5657
199, 200, 201, 202, 205, 209, 215 periphyton, 73, 74, 78, 102
Pallinup Siltstone, 264 Persian literature, 14, 46
palm, 32. See alsooil palm perturbations, 243249, 258, 264
Easter Island ramp, 247
indigenous, 285 spike, 247
Jubaea, 284, 285, 291295, 303307, step, 247
312314 Peru, 96, 107
17th-century remains, 306 pesticides, 155, 165
pulp, 304 cyanobacteria, 73
Palmer, J. Linton, 280 pests, 226, 233
palm products and diseases, cyle of, 233
fruits, 304 susceptibility of crops to, 233
honey, 304 Petn, 60, 64, 65, 76, 77, 103
juice, 304 Central, 54, 65
extraction, 308 collapse, 56
nuts, 135, 296 Petexbatn, 72, 103
thatch, 104 petroplinthite, 125
wood for boat construction, 305 Petrus de Crescentiis, 186, 196, 201, 202, 203,
palm tree vegetation 205, 206, 209
Easter Island, 283, 284 phaeozem. Seechernozem
reconstruction, 292 Pharaonic period, 227
palynological studies, 283, 287, 305, 307 Philometer Attalos, 192
Easter Island, 274 phosphate, 227, 239, 330, 331
Panini, 20 migration of, 62
Papua New Guinea, 233 rock, 11, 217
Papua region, 275 phosphorus, 10, 62, 104, 125, 140, 145, 161,
Paramardideva Candella, 27 222, 225, 228, 231, 235, 238, 241,
Parashara, 25, 26 266, 268
parchment, 177 as archeological tool, 62
parent material, 8, 9, 164 budget, 266
variation, 218 enriched particles, 266
Parker Earthwork, Ontario, 109, 110 enriched topsoil, 268
pastoralism, 18, 19, 26, 45, 134, 137, 138, 139 fixation, 62
pasture, 2, 19, 25, 26, 255 from sediments, 267
improvement, 254 input, 269
overuse, 157 in milpas, 73
Pataliputra, 24 load, 266
Paucarcolla-Juliaca plain, 106 need, 266
peanuts, 76 particulate, 266
pearl millet, 135 reduction, 268
peas, 30 sources and transport, 268
black-eyed, 142 testing, 62
peasant knowledge of soils, 59 photosynthesis, 244
359
Index
Witwatersrand, 162 Z
Woden Creek, 253, 254
Woden Station, 253 Zambia, 151, 152, 154, 162, 163
woodland, 227, 232, 252, 265 Zimbabwe, 150, 152, 153, 156, 157, 158, 159,
clearance, 236 161. See also Great Zimbabwe
coniferous, 216 Zimbabwean plateau, 139
conversion to grassland, 221 Zopotec, 54
deciduous, 216, 221
Easter Island, 284
evergreen broadleaf, 220
fertility loss, 217
Mesolithic, 221
reduction of leaching by, 220
regeneration, 232
removal, 221, 222
wood ash, 216, 217, 223, 225, 227
wool production, 237
World Soil Map, 129
worms, 244
X
Xenophon, 192, 193, 194, 196, 197, 206
Xochimilco, 97
Xuen Zang, 25
Xunantunich, 72
Y
Yajurvediya Maitrayani Samhita, 19
Yalahau, 73, 102
yam(s), 142, 280
Yangambi, 129
Yanhuitlan Formation, 67
Yaqui River, 109
Yaxa soil series, 64
Yconomia, 202
yield
classes, 30
crop, 28, 30, 40, 216
decrease due to erosion, 312
farming, 325
levels, 228
Yoruba, 149
Yucatn, 53, 56, 58, 59, 61, 65, 71, 73, 74, 94,
99, 102
northern, 56
northwestern, 64
soil chronosequence, 59